Experimental and Modeling Study of Methyl Cyclohexane Pyrolysis and Oxidation

Experimental and Modeling Study of Methyl Cyclohexane Pyrolysis and Oxidation

J. P. Orme, H. J. Curran,* and J. M. Simmie

Chemistry Department, National UniVersity of Ireland, Galway, Ireland

ReceiVed: August 5, 2005; In Final Form: September 29, 2005

Although the combustion chemistry of aliphatic hydrocarbons has been extensively documented, the oxidationof cyclic hydrocarbons has been studied to a much lesser extent. To provide a deeper understanding of thecombustion chemistry of naphthenes, the oxidation of methylcyclohexane was studied in a series of high-temperature shock tube experiments. Ignition delay times for a series of mixtures, of varying methylcyclohexane/oxygen equivalence ratios (φ ) 0.5, 1.0, 2.0), were measured over reflected shock temperatures of 1200-2100 K and reflected shock pressures of 1.0, 2.0, and 4.0 atm. A detailed chemical kinetic mechanism hasbeen assembled to simulate the shock tube results and flow reactor experiments, with good agreement observed.

Introduction

Much work has been carried out on the oxidation andpyrolysis of aliphatic hydrocarbons in a wide range of combus-tion environments. This research has provided an understandingof the kinetic pathways by which these fuels react under variousconditions and has allowed the assembly of detailed chemicalkinetic mechanisms1 with considerable predictive powers.

However, the combustion chemistry of cyclic hydrocarbonshas not been explored as extensively, with the possible exceptionof aromatics (benzene,2-16 toluene,17-24 and alkylbenzenes25-28

receiving particular attention of late), whereas many naphthenes(cycloalkanes) have been neglected. There have been somestudies on the oxidation of cyclohexane, cyclohexene, andcyclohexa-1,3-diene by Lemaire et al.29 and Ribaucour et al.30

and also the oxidation of cyclohexene by Dayma et al.31

Naphthenes are present in gasoline32 and other fuels, empha-sizing the need to provide a better understanding of their com-bustion properties. Some jet fuels, such as the JP-8 family, arecomposed mainly of monocyclic and bicyclic alkanes,33 and upto 20% of Jet-A fuel, the commercial equivalent of JP-8, is alsocomposed of cycloalkanes.34 In addition to its role as a tradi-tional fuel, methylcyclohexane (MCH) can provide a heat sinkof up to 2190 kJ kg-1 through an endothermic dehydrogenationreaction, forming toluene, thus enhancing the cooling of jet en-gines, although the formation of aromatics may augment soot pro-duction and have a deleterious effect on the combustor liner.35

It is vital to understand the influence of the naphthenestructure and the mixture composition on ignition delay timesto control burning rates in homogeneous charge compressionignition engines and the “knocking” phenomenon in sparkignition engines.36 MCH, with a research octane number of 7537

and a cetane number of 20, is one naphthene for which thereare very few experimental data available, and thus it is the focusof this work. Previous work includes a study of the unimoleculardecomposition of MCH, as well as other methyl substitutedcycloalkanes, using the technique of very low pressure pyrolysisat temperatures in the range 861-1218 K.38 A rate constant forthe dominant primary reaction channel of ring opening adjacentto the methyl substituent group of 2.51× 1016 exp(-82457 calmol-1/RT) s-1 was reported.

A high temperature (1050-1200 K) turbulent flow reactorstudy39 has been carried out on pure MCH and blends of MCH

with toluene (one of its dehydrogenation products). MCHpyrolysis and oxidation decay profiles were determined attemperatures of 1058, 1108, 1154, and 1192 K, at atmosphericpressure. Species profiles were obtained over a 70 ms periodfor MCH oxidation at 1160 K and for MCH pyrolysis at 1155K. Major products, for both oxidation and pyrolysis, includedmethane, ethylene, propene, buta-1,3-diene, and isoprene.Aromatics (benzene and toluene) were among the minorproducts as well as ethane, cyclopentadiene, cyclohexene,isobutene, allene, acetylene, and propyne.

Ignition delay times of MCH/O2 mixtures were recordedbehind incident shock waves by Hawthorn and Nixon,40 in thetemperature range 1200-1480 K, at equivalence ratios,φ, of0.1-2.1, and pressures of 0.61, 1.02, and 1.70 atm. Most ofthe mixtures used, with the exceptions ofφ ) 0.1 and 0.2, werevery dilute blends of MCH and O2 in 99% argon.

Granata et al.41 performed a kinetic modeling study of thepyrolysis and oxidation of naphthenes, which allowed a semide-tailed kinetic scheme to be expanded to incorporate thecombustion properties of cyclohexane initially and was furtherextended to include MCH. This expansion was accomplishedby adding the primary propagation reactions of cyclohexane andMCH to an existing mechanism, which had already beenvalidated for the pyrolysis, partial oxidation, and combustionof hydrocarbons fuels such asn-heptane and isooctane.42-44

Experimental species mole fractions from the turbulent flowreactor study39 were used to verify the accuracy of the MCHmechanism, which was generally in good agreement with theexperimental data. Moreover, this mechanism was used tosimulate species profiles in laminar premixed flames for so-called surrogate mixtures of pure compounds that represent theaviation fuel JP-8.34

Specific Objectives

The aim of this study is to provide more information on thecombustion characteristics of MCH and to develop a detailedchemical kinetic mechanism to enhance our knowledge of thepyrolysis and oxidation processes of naphthene fuels.

Experimental Section

The stainless steel shock tube consists of a large but short(52 cm diameter and 63 cm long) driver section coupled via a

114 J. Phys. Chem. A2006,110,114-131

10.1021/jp0543678 CCC: $33.50 © 2006 American Chemical SocietyPublished on Web 12/13/2005

10 cm long transition piece to the test section that is 622 cmlong and 10.24 cm internal diameter. Pressure transducers (PCBPiezotronics, model 113A21) were set into the final 50 cm ofthe test section and were used to measure the incident shockvelocity with the aid of two universal time counters (HP5300A,TTi Apollo 100) and an oscilloscope (Tektronix TDS 2024).To allow for shock attenuation, the shock velocity at the endwallwas calculated by extrapolating the incident velocities to theendwall. Reflected shock conditions were calculated, using theusual one-dimensional shock relations45 and the applicationGasEq,46 from initial temperatures in the range 290-298 K,and initial pressures in the range 12-107 Torr.

A polycarbonate diaphragm was allowed to burst underpressure with the assistance of a cross shaped cutting devicethat petaled the diaphragm. This shock tube was fully character-ized47 in line with a previous account48 and validated againstrecent experiments onn-heptane oxidation from the Stanfordgroup.49

Mixture Preparation. Test mixtures were prepared in a 50L stainless steel tank using standard manometric methods. Gaseswere obtained from BOC Ireland Ltd.; helium CP Grade99.999%, argon Zero Grade 99.998%, and oxygen ResearchGrade 99.985%. All gases were used without further purification.MCH was obtained from Aldrich Chemical Co. Ltd. and wasdetermined to be 99.6% pure by GC analysis. To minimize thepresence of atmospheric air in the sample, the liquid MCH wassubjected to several freeze-pump-thaw degassing cycles beforebeing used. The liquid MCH was incorporated into the mixingvessel by vaporization into the evacuated (10-6 Torr) mixingtank, and partial pressures were measured using a 100 TorrBaratron gauge to an accuracy of 0.01 Torr. Argon was addedusing a Wallace and Tiernan 800 Torr absolute pressure gauge.The exact composition of mixtures used during this study areshown in Table 1. Test gas mixtures were normally made up toa final pressure of 800 Torr and allowed to stand for 24 h toensure homogeneity. From the resulting mix, initial pressures,p1, varying from 12 to 107 Torr were used.

To test the consistency of this method, experiments werecarried out on identical test gas mixtures that had been leftstanding for various lengths of time. There was no variation inthe data produced using these mixtures.

Experimental Procedure.Prior to each experiment both thedriver and driven sections of the shock tube were evacuatedindependently. The driver section was evacuated to 10-3 Torrusing an Edwards oil rotary pump, whereas the driven sectionwas first pumped to 10-3 Torr using an Edwards oil rotarypump, and the final pressure of 10-6 Torr was achieved usingan Edwards oil diffusion pump.

Ignition Delay Time Measurements.Emissions were ob-served using an end on detection diagnostic that consisted of aPDA55 (switchable gain, amplified silicon detector) locatedbehind a 431 nm (for CH* emission) narrow band-pass filterwith a spectral bandwidth of 10 nm. The filter and the PDAwere aligned behind a quartz window located in the endwall.

A Kistler pressure transducer, mounted flush with the endwall,signaled the shock wave arrival at the endwall, Figure 1, andthe beginning of the ignition delay period. The end of the periodwas defined as the maximum rise in the rate of emission.

A comparative study was carried out by repeating a series ofshock tube experiments performed by Hawthorn and Nixon,40

who defined ignition delay times as the length of time betweenshock arrival at the measurement plane, as indicated by apressure transducer, and the first emission of light detected atthe measurement plane. For these experiments the same end-on detection diagnostic was used, but the filter was removed tobe consistent with the Hawthorn and Nixon study.

Simulated Ignition Delay Times and Species Profiles.TheMCH oxidation mechanism was used, in conjunction with theChemkin III suite of programs,51 to simulate both the shocktube ignition delay times produced in this study and the flowreactor species profiles produced by Zeppieri et al.39 Under theassumptions of constant pressure, uniform mixing and infinitelyfast mass transport to the walls within the turbulent flow reactor,the Aurora module of Chemkin III was used to simulate thebehavior of MCH under the experimental conditions of Zeppieriet al. The Shock module was used, withp5 andT5 supplied tothe reflected shock mode, to simulate the ignition delaysobserved in this study.

For experimental ignition delays the end of the delay timewas defined as the maximum rate of change of CH* emission.However, the mechanism used in this study, in common withmost other mechanisms, does not contain the electronicallyexcited CH* species. Horning et al.50 states that the rate ofemission can be inferred from the rate of production of CH*:

and defines the simulated ignition delay time as the time atwhich the maximum value of [C˙ 2H][O] occurs. This definitionwas used in this study.

Mechanism Formulation

The MCH mechanism is based on the reaction schemedeveloped by Laskin et al.,52 which describes the oxidation ofbuta-1,3-diene, although the H2/O2 submechanism has beenreplaced with that recently published by OÄ Conaire et al.53 Tothis has been added the detailed chemistry that describes MCHoxidation and the oxidation of other intermediate hydrocarbonspecies larger than buta-1,3-diene. Rate constants for MCHoxidation are based on the analysis of Curran et al. forn-heptane54 and isooctane55 oxidation, in which a detaileddescription of reaction rate rules are described. Here we provide

TABLE 1: Percent Composition of MCH/O2/Ar Mixtures

φ MCH O2 Ar

0.5 0.5 10.50 89.001.0 1.0 10.50 88.502.0 1.0 5.25 93.750.105 0.1 9.90 90.00

Figure 1. Typical endwall pressure sensor (1) and CH* emission (2)profiles. Conditions: 1.0% MCH, 10.5% O2, 88.5% Ar, 1327.5 K, 2.0atm.

C2H + O f CH* + CO

Methyl Cyclohexane Pyrolysis and Oxidation J. Phys. Chem. A, Vol. 110, No. 1, 2006115

only details of updated rate constant expressions relative to thosepublished for isooctane. Because so many rate expressions forMCH and its radical species have been derived using micro-scopic reversibility, the thermodynamic parameters of thesespecies are included, Table 2; the thermodynamic values forMCH used in this study are in good agreement with previouswork by Beckett et al.,57 Table 2. These values have beendetermined using THERM58 based on Benson’s group additivityestimation method.59 H/C/O groups and bond dissociationgroups have been updated by taking into account the work ofLay et al.60 and Sumathi and Green.61 As the hydrogen/oxygen53

and the buta-1,3-diene52 submechanisms already appear in theliterature, we have included only the methylcyclohexane sub-mechanism, Table 3, whereA is the pre-exponential frequencyfactor,n is the temperature curvature dependence, andEa is theactivation energy for reaction. A full listing of the detailedchemical kinetic mechanism is available in Chemkin formatfrom the authors ([email protected]).

Naming Scheme.Species names were created for use inTable 3 and Chemkin. To convert Table 3 to Chemkin format,an “X” was added before all species whose name began with anumber and all commas and dashes were removed. Most specieswere given the notation “number-molecular formula-number”or “letter-molecular formula-letter”. The numbers and letterscan be explained as follows:

•For straight chain hydrocarbons the carbons were labeled witha number, starting with C1 upward along the chain, Figure 2.

•For branched hydrocarbons the carbons were labeled with aletter. For example, for 2-methylhexane the first carbon atomin the chain was labeled with an “a”, the second with a “b” and

so on, Figure 3. For 3-methylhexane the first atom was labeledwith a “g” and so on, Figure 4.

•Where double bonds were present, numbers or letters wereadded before the molecular formula to denote their position onthe hydrocarbon.

•Where radical sites were present, numbers or letters wereadded after the formula to denote their positions as shown,Figures 2-4.

Cyclic hydrocarbons and some branched species were givenunique names that do not conform to the above naming schemes;these species are illustrated and named in Figure 5.

Species that were already present in the reaction schemedeveloped by Laskin et al., and the H2/O2 submechanism by OÄConaire, were not renamed, with the exceptions of buta-1,3-diene, but-1-en-4-yl, and benzene, whose names were changedfrom C4H6 to 1,3C4H6, C4H7 to 1C4H7-4, and A1 to C6H6,respectively.

Unimolecular Fuel Decomposition.MCH decompositioncan initiate through bond homolysis on the six-membered ring,leading to ring opening and the formation of C7H14 diradicalspecies. Bond homolysis can occur at six different sites withthree unique diradicals being formed; homolysis of bonds thatare equidistant from the methyl substituent lead to formationof equivalent species. The rate constant expressions used in thisstudy for reactions of this type were consistent with thosepublished by Brown and King in a very low pressure pyrolysisstudy of methyl- and ethynyl- cyclopentanes and cyclohexanes,

TABLE 2: Thermodynamic Properties for Selected Species

Cp

species ∆fH°(298 K) S°(298 K) 300 K 400 K 500 K 600 K 800 K 1000 K 1500 K

MCH -36.75 81.64 32.67 44.46 54.95 63.77 77.32 87.02 101.50MCHa -36.99 82.05 32.51 44.35 55.21 64.46 78.74 88.79 103.20MCH-R1 7.65 86.88 31.89 41.98 51.40 59.62 72.57 82.00 96.11MCH-R2 9.60 86.08 31.17 42.13 51.85 60.38 73.57 82.57 96.30MCH-R3 9.60 86.08 31.17 42.13 51.85 60.38 73.57 82.57 96.30MCH-R4 9.60 86.08 31.17 42.13 51.85 60.38 73.57 82.57 96.30CYCHEXCH2 12.25 85.05 31.90 43.10 53.04 61.37 74.16 83.28 96.84

a MCH: ∆fH° (kcal/mol) values from Prosen et al.,56 S° [cal/(mol K)] andCp [cal/(mol K)] values from Beckett et al.57

Figure 2. Naming scheme for straight chained hydrocarbons.

Figure 3. Naming scheme for branched hydrocarbons with a methylgroup attached to the second carbon atom.

Figure 4. Naming scheme for branched hydrocarbons with a methylgroup attached to the third carbon atom.

Figure 5. Names of miscellaneous hydrocarbons added to the MCHmechanism.

116 J. Phys. Chem. A, Vol. 110, No. 1, 2006 Orme et al.

TABLE 3: Methylcyclohexane Mechanism Rate Coefficients; cm3/mol/s/cal Units

reaction A n EA reaction A n EA

1. MCH ) C7H14-g,l 2.5E+16 0.0 86042 68. MCH+ O ) MCH-R4 + OH 5.51E+05 2.5 28302. MCH ) C7H14-a,f 2.5E+16 0.0 86042 69. MCH+ O ) CYCHEXCH2+ OH 9.80E+05 2.4 47503. MCH ) C7H14-1,6 2.5E+16 0.0 82457 70. 1,6C7H12 + H ) 1,6C7H11-4 + H2 1.30E+06 2.4 44714. CYCHEXRAD+ CH3 ) MCH 6.63E+14 -0.6 0 71. 1,6C7H12 + CH3 ) 1,6C7H11-4 + CH4 1.51E+00 3.5 54815. 1C6H11-6 ) CYCHEXRAD 1.00E+08 0.9 5900 72. 1,6C7H12 + O ) 1,6C7H11-4 + OH 5.51E+05 2.5 28306. aC7H13-f ) MCH-R1 1.00E+08 0.9 5900 73. 1,6C7H12 + OH ) 1,6C7H11-4 + H2O 4.67E+07 1.6 -357. gC7H13-l ) MCH-R2 1.00E+08 0.9 5900 74. iC5H8 + H ) iC5H7 + H2 1.73E+05 2.5 24928. 2C7H13-7 ) MCH-R2 1.00E+08 0.9 8700 75. iC5H8 + CH3 ) iC5H7 + CH4 2.21E+00 3.5 56759. 1C7H13-6 ) MCH-R3 1.00E+08 0.9 8500 76. iC5H8 + OH ) iC5H7 + H2O 3.12E+06 2.0 -298

10. kC7H13-g ) MCH-R3 1.00E+08 0.9 5900 77. iC5H8 + HO2 ) iC5H7 + H2O2 9.64E+03 2.6 1391011. eC7H13-a ) MCH-R4 1.00E+08 0.9 5900 78. iC5H8 + CH3O ) iC5H7 + CH3OH 9.00E+01 3.0 1198712. 1,3C6H9-6 ) cycC6H9-3 1.00E+08 0.9 6900 79. iC5H8 + O ) iC5H7 + OH 4.41E+05 2.4 315013. 1,4C6H9-6 ) cycC6H9-4 1.00E+08 0.9 6900 80. iC5H8 + O2 ) iC5H7 + HO2 3.30E+12 0.0 3990014. 1C7H13-7 ) CYCHEXCH2 1.00E+08 0.9 1903 81. CYCHEXENE+ H ) cycC6H9-3 + H2 6.75E+05 2.4 20715. 2C6H10-1,6) CYCHEXENE 1.00E+08 0.9 1000 82. CYCHEXENE+ CH3 ) cycC6H9-3 + CH4 7.38E+00 3.3 400216. 3C6H10-1,6) CYCHEXENE 1.00E+08 0.9 1000 83. CYCHEXENE+ O ) cycC6H9-3 + OH 1.32E+06 2.4 121017. aC7H13-f ) aC7H13-c 3.67E+12 -0.6 15300 84. CYCHEXENE+ OH ) cycC6H9-3 + H2O 5.53E+04 2.6 -191918. 1C7H13-7 ) 1C7H13-3 2.80E+10 0.0 7500 85. CYCHEXENE+ H ) cycC6H9-4 + H2 2.60E+06 2.4 447119. gC7H13-l ) gC7H13-i 3.67E+12 -0.6 13000 86. CYCHEXENE+ CH3 ) cycC6H9-4 + CH4 3.02E+00 3.5 548120. 2C7H13-7 ) 2C7H13-4 3.67E+12 -0.6 15300 87. CYCHEXENE+ O ) cycC6H9-4 + OH 1.10E+06 2.5 283021. 2C7H13-7 ) 2C7H13-1 2.80E+10 0.0 31640 88. CYCHEXENE+ OH ) cycC6H9-4 + H2O 9.34E+07 1.6 -3522. 1C7H13-6 ) 1C7H13-3 3.67E+12 -0.6 17900 89. a,eC7H12 + H ) a,eC7H11-c + H2 3.38E+05 2.4 20723. kC7H13-g ) kC7H13-j 3.67E+12 -0.6 15300 90. a,eC7H12 + H ) a,eC7H11-d + H2 3.38E+05 2.4 20724. eC7H13-a ) eC7H13-d 3.67E+12 -0.6 15300 91. 1,5C7H12 + H ) 1,5C7H11-3 + H2 3.38E+05 2.4 20725. 1C6H11-6 ) 1C6H11-3 3.67E+12 -0.6 15300 92. 1,5C7H12 + H ) 1,5C7H11-4 + H2 3.38E+05 2.4 20726. C7H14-1,6) 1C7H14 3.67E+12 -0.6 15300 93. g,kC7H12 + H ) g,kC7H11-j + H2 3.38E+05 2.4 207

declared duplicate reaction... 94. 1,4C5H8 + H ) 1,4C5H7-3 + H2 3.38E+05 2.4 20727. C7H14-1,6) 1C7H14 2.80E+10 0.0 25875 95. a,eC7H12 + CH3 ) a,eC7H11-c + CH4 3.69E+00 3.3 4002

declared duplicate reaction... 96. a,eC7H12 + CH3 ) a,eC7H11-d + CH4 3.69E+00 3.3 400228. C7H14-1,6) 2C7H14 3.67E+12 -0.6 12800 97. 1,5C7H12 + CH3 ) 1,5C7H11-3 + CH4 3.69E+00 3.3 400229. C7H14-a,f ) aC7H14 3.67E+12 -0.6 10500 98. 1,5C7H12 + CH3 ) 1,5C7H11-4 + CH4 3.69E+00 3.3 400230. C7H14-a,f ) eC7H14 3.67E+12 -0.6 12800 99. g,kC7H12 + CH3 ) g,kC7H11-j + CH4 3.69E+00 3.3 400231. C7H14-g,l ) gC7H14 3.67E+12 -0.6 12800 100. 1,4C5H8 + CH3 ) 1,4C5H7-3 + CH4 3.69E+00 3.3 400232. C7H14-g,l ) kC7H14 3.67E+12 -0.6 12800 101. a,eC7H12 + O ) a,eC7H11-c + OH 6.60E+05 2.4 121033. 2C5H9-5 ) 2C5H9-1 3.67E+12 -0.6 10000 102. a,eC7H12 + O ) a,eC7H11-d + OH 6.60E+05 2.4 121034. 1C5H9-4 ) 1C5H9-3 3.56E+10 0.9 31500 103. 1,5C7H12 + O ) 1,5C7H11-3 + OH 6.60E+05 2.4 121035. MCH+ H ) MCH-R1 + H2 6.02E+05 2.4 2583 104. 1,5C7H12 + O ) 1,5C7H11-4 + OH 6.60E+05 2.4 121036. MCH+ H ) MCH-R2 + H2 2.60E+06 2.4 4471 105. g,kC7H12 + O ) g,kC7H11-j + OH 6.60E+05 2.4 121037. MCH+ H ) MCH-R3 + H2 2.60E+06 2.4 4471 106. 1,4C5H8 + O ) 1,4C5H7-3 + OH 6.60E+05 2.4 121038. MCH+ H ) MCH-R4 + H2 1.30E+06 2.4 4471 107. a,eC7H12 + OH ) a,eC7H11-c + H2O 2.76E+04 2.6 -191939. MCH+ H ) CYCHEXCH2+ H2 6.65E+05 2.5 6756 108. a,eC7H12 + OH ) a,eC7H11-d + H2O 2.76E+04 2.6 -191940. MCH+ CH3 ) MCH-R1 + CH4 6.01E-10 6.4 893 109. 1,5C7H12 + OH ) 1,5C7H11-3 + H2O 2.76E+04 2.6 -191941. MCH+ CH3 ) MCH-R2 + CH4 3.02E+00 3.5 5481 110. 1,5C7H12 + OH ) 1,5C7H11-4 + H2O 2.76E+04 2.6 -191942. MCH+ CH3 ) MCH-R3 + CH4 3.02E+00 3.5 5481 111. g,kC7H12 + OH ) g,kC7H11-j + H2O 2.76E+04 2.6 -191943. MCH+ CH3 ) MCH-R4 + CH4 1.51E+00 3.5 5481 112. 1,4C5H8 + OH ) 1,4C5H7-3 + H2O 2.76E+04 2.6 -191944. MCH+ CH3 ) CYCHEXCH2+ CH4 4.52E-01 3.6 7154 113. 1,5C6H10 + H ) 1,5C6H9-3 + H2 6.75E+05 2.4 20745. MCH+ O2 ) MCH-R1 + HO2 1.00E+13 0.0 48200 114. 1,5C6H10 + CH3 ) 1,5C6H9-3 + CH4 7.38E+00 3.3 400246. MCH+ O2 ) MCH-R2 + HO2 4.00E+13 0.0 49640 115. 1,5C6H10 + O ) 1,5C6H9-3 + OH 1.32E+06 2.4 121047. MCH+ O2 ) MCH-R3 + HO2 4.00E+13 0.0 49640 116. 1,5C6H10 + OH ) 1,5C6H9-3 + H2O 5.53E+04 2.6 -191948. MCH+ O2 ) MCH-R4 + HO2 2.00E+13 0.0 49640 117. 1,6C7H12 + H ) 1,6C7H11-3 + H2 6.75E+05 2.4 20749. MCH+ O2 ) CYCHEXCH2+ HO2 3.00E+13 0.0 52290 118. 1,6C7H12 + CH3 ) 1,6C7H11-3 + CH4 7.38E+00 3.3 400250. MCH+ HO2 ) MCH-R1 + H2O2 3.61E+03 2.5 10532 119. 1,6C7H12 + O ) 1,6C7H11-3 + OH 1.32E+06 2.4 121051. MCH+ HO2 ) MCH-R2 + H2O2 1.93E+04 2.6 13910 120. 1,6C7H12 + OH ) 1,6C7H11-3 + H2O 5.53E+04 2.6 -191952. MCH+ HO2 ) MCH-R3 + H2O2 1.93E+04 2.6 13910 121. 1,5C7H12 + H ) 1,5C7H11-7 + H2 1.73E+05 2.5 249253. MCH+ HO2 ) MCH-R4 + H2O2 9.64E+03 2.6 13910 122. 1,3C5H8 + H ) 1,3C5H75 + H2 1.73E+05 2.5 249254. MCH+ HO2 ) CYCHEXCH2+ H2O2 2.38E+04 2.5 16494 123. a,eC7H12 + H ) a,eC7H11-a2+ H2 1.73E+05 2.5 249255. MCH+ OH ) MCH-R1 + H2O 5.73E+10 0.5 63 124. 1,5C7H12 + CH3 ) 1,5C7H11-7 + CH4 2.21E+00 3.5 567556. MCH+ OH ) MCH-R2 + H2O 9.34E+07 1.6 -35 125. 1,3C5H8 + CH3 ) 1,3C5H7-5 + CH4 2.21E+00 3.5 567557. MCH+ OH ) MCH-R3 + H2O 9.34E+07 1.6 -35 126. a,eC7H12 + CH3 ) a,eC7H11-a2+ CH4 2.21E+00 3.5 567558. MCH+ OH ) MCH-R4 + H2O 4.67E+07 1.6 -35 127. 1,5C7H12 + O ) 1,5C7H11-7 + OH 4.41E+05 2.4 315059. MCH+ OH ) CYCHEXCH2+ H2O 5.27E+09 1.0 1586 128. 1,3C5H8 + O ) 1,3C5H7-5 + OH 4.41E+05 2.4 315060. MCH+ CH3O ) MCH-R1 + CH3OH 2.29E+10 0.0 2873 129. a,eC7H12 + O ) a,eC7H11-a2+ OH 4.41E+05 2.4 315061. MCH+ CH3O ) MCH-R2 + CH3OH 2.90E+11 0.0 4571 130. 1,5C7H12 + OH ) 1,5C7H11-7 + H2O 3.12E+06 2.0 -29862. MCH+ CH3O ) MCH-R3 + CH3OH 2.90E+11 0.0 4571 131. 1,3C5H8 + OH ) 1,3C5H7-5 + H2O 3.12E+06 2.0 -29863. MCH+ CH3O ) MCH-R4 + CH3OH 1.45E+11 0.0 4571 132. a,eC7H12 + OH ) a,eC7H11-a2+ H2O 3.12E+06 2.0 -29864. MCH+ CH3O ) CYCHEXCH2+ CH3OH 2.17E+11 0.0 6458 133. g,kC7H12 + H ) g,kC7H11-m + H2 6.65E+05 2.5 675665. MCH+ O ) MCH-R1 + OH 3.83E+05 2.4 1140 134. g,kC7H12 + CH3 ) g,kC7H11-m + CH4 4.52E-01 3.6 715466. MCH+ O ) MCH-R2 + OH 1.10E+06 2.5 2830 135. g,kC7H12 + O ) g,kC7H11-m + OH 9.80E+05 2.4 475067. MCH+ O ) MCH-R3 + OH 1.10E+06 2.5 2830 136. g,kC7H12 + OH ) g,kC7H11-m + H2O 5.27E+09 1.0 1586


TABLE 3: Continued


137. g,kC7H12 + H ) g,kC7H11-i + H2 2.65E+06 2.2 0 205. MCH3ENE+ H ) MCH-R4 1.32E+13 0.0 3260138. g,kC7H12 + CH3 ) g,kC7H11-i + CH4 4.61E+00 3.1 2330 206. 1,3C6H10 + H ) 1C6H11-3 1.32E+13 0.0 3260139. g,kC7H12 + OH ) g,kC7H11-i + H2O 6.14E+02 3.2 -3500 207. a,cC7H12 + H ) aC7H13-c 1.32E+13 0.0 3260140. 1C7H14 + H ) 1C7H13-3 + H2 3.38E+05 2.4 207 208. 1,3C7H12 + H ) 1C7H13-3 1.32E+13 0.0 3260141. 1C7H14 + CH3 ) 1C7H13-3 + CH4 3.69E+00 3.3 4002 209. g,iC7H12 + H ) gC7H13-i 1.32E+13 0.0 3260142. 1C7H14 + OH ) 1C7H13-3 + H2O 2.76E+04 2.6 -1919 210. 2,4C7H12 + H ) 2C7H13-4 1.32E+13 0.0 3260143. 1C7H14 + HO2 ) 1C7H13-3 + H2O2 4.82E+03 2.5 10532 211. c,eC7H12 + H ) eC7H13-d 1.32E+13 0.0 3260144. 1C7H14 + CH3O ) 1C7H13-3 + CH3OH 4.00E+01 2.9 8609 212. 1,3CYCHEX+ H ) cycC6H9-4 1.32E+13 0.0 3260145. 1C7H14 + O ) 1C7H13-3 + OH 6.60E+05 2.4 1210 213. 1,4CYCHEX+ H ) cycC6H9-4 1.32E+13 0.0 3260146. 1C7H14 + O2 ) 1C7H13-3 + HO2 2.20E+12 0.0 37220 214. 1,3C5H8 + H ) 2C5H9-1 1.32E+13 0.0 3260147. 2C7H14 + H ) 2C7H13-1 + H2 1.73E+05 2.5 2492 215. aC7H13-f + H ) C7H14-a,f 1.32E+13 0.0 4660148. 2C7H14 + H ) 2C7H13-4 + H2 3.38E+05 2.4 207 216. a,eC7H12 + H ) eC7H13-a 1.32E+13 0.0 4660149. 2C7H14 + CH3 ) 2C7H13-1 + CH4 4.52E-01 3.6 7154 217. iC5H8 + H ) cC5H9-a 1.32E+13 0.0 4660150. 2C7H14 + CH3 ) 2C7H13-4 + CH4 3.69E+00 3.3 4002 218. g,k,mC7H10 + H ) g,kC7H11-m 1.32E+13 0.0 4660151. 2C7H14 + O ) 2C7H13-1 + OH 4.41E+05 2.4 3150 219. g,i,kC7H10 + H ) g,kC7H11-j 1.32E+13 0.0 4660152. 2C7H14 + OH ) 2C7H13-1 + H2O 3.12E+06 2.0 -298 220. METCYCHEX+ H ) CYCHEXCH2 1.32E+13 0.0 4660153. 2C7H14 + O ) 2C7H13-4 + OH 6.60E+05 2.4 1210 221. MCH1ENE+ H ) MCH-R2 1.32E+13 0.0 4660154. 2C7H14 + OH ) 2C7H13-4 + H2O 2.76E+04 2.6 -1919 222. i,kC7H12 + H ) kC7H13-j 1.32E+13 0.0 4660155. aC7H14 + H ) aC7H13-c + H2 3.38E+05 2.4 207 223. 1C7H13-7 + H ) C7H14-1,6 2.64E+13 0.0 2160156. aC7H14 + CH3 ) aC7H13-c + CH4 3.69E+00 3.3 4002 224. 1,6C7H12 + H ) 1C7H13-6 2.64E+13 0.0 2160157. aC7H14 + O ) aC7H13-c + OH 6.60E+05 2.4 1210 225. 1,3C5H8 + H ) 1C5H9-4 2.64E+13 0.0 2160158. aC7H14 + OH ) aC7H13-c + H2O 2.76E+04 2.6 -1919 226. g,k,mC7H10 + H ) g,k-C7H11-i 2.64E+13 0.0 2160159. eC7H14 + H ) eC7H13-d + H2 3.38E+05 2.4 207 227. METCYCHEX+ H ) MCH-R1 2.64E+13 0.0 2160160. eC7H14 + CH3 ) eC7H13-d + CH4 3.69E+00 3.3 4002 228. g,mC7H12 + H ) gC7H13-i 2.64E+13 0.0 2160161. eC7H14 + O ) eC7H13-d + OH 6.60E+05 2.4 1210 229. 1C4H7-4 + C2H4 ) 1C6H11-6 1.32E+04 2.5 6130162. eC7H14 + OH ) eC7H13-d + H2O 2.76E+04 2.6 -1919 230. 2C5H9-5 + C2H4 ) 2C7H13-7 1.32E+04 2.5 6130163. gC7H14 + H ) gC7H13-i + H2 2.65E+06 2.2 0 231. 1C5H9-5 + C2H4 ) 1C7H13-7 1.32E+04 2.5 6130164. gC7H14 + CH3 ) gC7H13-i + CH4 4.61E+00 3.1 2330 232. aC5H9-d + C2H4 ) aC7H13-f 1.32E+04 2.5 6130165. gC7H14 + OH ) gC7H13-i + H2O 6.14E+02 3.2 -3500 233. cC5H9-a + C2H4 ) gC7H13-l 1.32E+04 2.5 6130166. kC7H14 + H ) kC7H13-j + H2 3.38E+05 2.4 207 234. C2H4 + aC3H5 ) 1C5H9-5 1.32E+04 2.5 6130167. kC7H14 + CH3 ) kC7H13-j + CH4 3.69E+00 3.3 4002 235. C2H4 + sC3H5 ) 2C5H9-5 1.32E+04 2.5 6130168. kC7H14 + O ) kC7H13-j + OH 6.60E+05 2.4 1210 236. C2H4 + tC3H5 ) aC5H9-d 1.32E+04 2.5 6130169. kC7H14 + OH ) kC7H13-j + H2O 2.76E+04 2.6 -1919 237. C2H4 + n-C4H5 ) 1,3C6H9-6 1.32E+04 2.5 6130170. CYCHEXENE+ H ) CYCHEXRAD 1.32E+13 0.0 3260 238. 1C5H9-4 + C2H4 ) kC7H13-g 1.32E+04 2.5 6130171. 1,5C6H10 + H ) 1C6H11-6 1.32E+13 0.0 3260 239. C5H10-1,4+ C2H4 ) C7H14-1,6 1.32E+04 2.5 6130172. 1C7H13-6 + H ) C7H14-1,6 1.32E+13 0.0 3260 240. C5H10-a,d+ C2H4 ) C7H14-a,f 1.32E+04 2.5 6130173. 2C7H13-7 + H ) C7H14-1,6 1.32E+13 0.0 3260 241. C5H10-1,4+ C2H4 ) C7H14-g,l 1.32E+04 2.5 6130174. 1,5C7H12 + H ) 1C7H13-6 1.32E+13 0.0 3260 242. C5H10-a,d+ C2H4 ) C7H14-g,l 1.32E+04 2.5 6130175. 1,5C7H12 + H ) 2C7H13-7 1.32E+13 0.0 3260 243. C3H6 + 1C4H7-4 ) 1C7H13-6 8.80E+03 2.5 6130176. 1,6C7H12 + H ) 1C7H13-7 1.32E+13 0.0 3260 244. 1C4H7-4 + C3H6 ) eC7H13-a 8.80E+03 2.5 6130177. eC7H13-a + H ) C7H14-a,f 1.32E+13 0.0 3260 245. C3H6 + C2H3 ) 1C5H9-4 8.80E+03 2.5 6130178. a,eC7H12 + H ) aC7H13-f 1.32E+13 0.0 3260 246. 1,3C4H6 + aC3H5 ) 1,5C7H11-7 8.80E+03 2.5 6130179. gC7H13-l + H ) C7H14-g,l 1.32E+13 0.0 3260 247. 1,3C4H6 + C2H5 ) 1C6H11-3 8.80E+03 2.5 6130180. kC7H13-g + H ) C7H14-g,l 1.32E+13 0.0 3260 248. iC5H8 + C2H5 ) aC7H13-c 8.80E+03 2.5 6130181. iC5H8 + H ) aC5H9-d 1.32E+13 0.0 3260 249. 1,3C4H6 + nC3H7 ) 1C7H13-3 8.80E+03 2.5 6130182. g,kC7H12 + H ) gC7H13-l 1.32E+13 0.0 3260 250. iC5H8 + C2H5 ) gC7H13-i 8.80E+03 2.5 6130183. g,kC7H12 + H ) kC7H13-g 1.32E+13 0.0 3260 251. 1,3C5H8 + C2H5 ) 2C7H13-4 8.80E+03 2.5 6130184. 1,4C5H8 + H ) 1C5H9-5 1.32E+13 0.0 3260 252. tC3H5 + 1,3C4H6 ) a,eC7H11-d 8.80E+03 2.5 6130185. 1,4C5H8 + H ) 1C5H9-4 1.32E+13 0.0 3260 253. 1,3C4H6 + iC3H7 ) eC7H13-d 8.80E+03 2.5 6130186. 1,3C5H8 + H ) 2C5H9-5 1.32E+13 0.0 3260 254. 1,3C4H6 + C2H3 ) 1,4C6H9-6 8.80E+03 2.5 6130187. C3H6 + C2H3 ) cC5H9-a 1.32E+13 0.0 3260 255. 1,3C4H6 + C2H3 ) 1,5C6H9-3 8.80E+03 2.5 6130188. 1,3,6C7H10 + H ) 1,6C7H11-3 1.32E+13 0.0 3260 256. 1,4C5H8 + C2H3 ) 1,6C7H11-4 8.80E+03 2.5 6130189. 1,3,5C6H8 + H ) 1,5C6H9-3 1.32E+13 0.0 3260 257. 1,3C4H6 + sC3H5 ) 1,5C7H11-3 8.80E+03 2.5 6130190. 1,3,6C7H10 + H ) 1,6C7H11-4 1.32E+13 0.0 3260 258. C2H3 + 1,3C5H8 ) 1,5C7H11-4 8.80E+03 2.5 6130191. 1,3,5C7H10 + H ) 1,5C7H11-3 1.32E+13 0.0 3260 259. C2H3 + 1,4-C5H8 ) g,kC7H11-m 8.80E+03 2.5 6130192. 1,3,5C7H10 + H ) 1,5C7H11-4 1.32E+13 0.0 3260 260. iC5H8 + C2H3 ) a,eC7H11-c 8.80E+03 2.5 6130193. 1,3,6C7H10 + H ) 1,5C7H11-7 1.32E+13 0.0 3260 261. 1,3C4H6 + aC3H5 ) 1,6C7H11-4 8.80E+03 2.5 6130194. a,c,eC7H10 + H ) a,eC7H11-c 1.32E+13 0.0 3260 262. 1,3C4H6 + aC3H5 ) 1,6C7H11-3 8.80E+03 2.5 6130195. a,c,eC7H10 + H ) a,eC7H11-d 1.32E+13 0.0 3260 263. 1,3C4H6 + aC3H5 ) g,kC7H11-m 8.80E+03 2.5 6130196. 1,3CYCHEX+ H ) cycC6H9-3 1.32E+13 0.0 3260 264. 1,3C4H6 + aC3H5 ) g,kC7H11-i 8.80E+03 2.5 6130197. 1,3,5C6H8 + H ) 1,3C6H9-6 1.32E+13 0.0 3260 265. C4H8-1,4+ C3H6 ) C7H14-1,6 8.80E+03 2.5 6130198. 1,3,5C6H8 + H ) 1,4C6H9-6 1.32E+13 0.0 3260 266. 1,3C4H6 + nC3H7 ) 2C7H13-1 8.80E+03 2.5 6130199. 1,4C6H9-6 + H ) 2C6H10-1,6 1.32E+13 0.0 3260 267. 1C6H11-6 + CH3 ) C7H14-a,f 1.89E+03 2.7 6850200. 1,3C6H9-6 + H ) 3C6H10-1,6 1.32E+13 0.0 3260 268. 1,5C6H10 + CH3 ) eC7H13-a 1.89E+03 2.7 6850201. MCH1ENE+ H ) MCH-R1 1.32E+13 0.0 3260 269. 1,3C4H6 + CH3 ) cC5H9-a 1.89E+03 2.7 6850202. MCH2ENE+ H ) MCH-R2 1.32E+13 0.0 3260 270. 1,3,5C6H8 + CH3 ) g,kC7H11-j 1.89E+03 2.7 6850203. MCH2ENE+ H ) MCH-R3 1.32E+13 0.0 3260 271. CYCHEXENE+ CH3 ) MCH-R2 1.89E+03 2.7 6850204. MCH3ENE+ H ) MCH-R3 1.32E+13 0.0 3260 272. 1,3C6H10 + CH3 ) kC7H13-j 1.89E+03 2.7 6850


Table 4.38 The authors estimated rate expressions and comparedsimulated MCH decomposition profiles to their experimentalraw data with satisfactory agreement. Their estimations werebased on known values for the corresponding carbon carbonbond fissions in open chain alkanes and are in good agreementwith the literature.

The reaction forming a cyclohexyl (CYCHEXRAD), Figure5, and a methyl radical was determined using the reverserecombination rate constant of 6.63× 1014T-0.57 cm3 mol-1

s-1 taken from the work of Tsang63 for methyl addition to anisopropyl radical and then calculated using microscopic revers-ibility.

Hydrogen Atom Abstraction. Hydrogen atom abstractionfrom MCH takes place at primary, secondary, and tertiary sites.Rate constant expressions for site-specific hydrogen atomabstraction reactions by various radicals have been publishedpreviously.54,55,64However, these values are continually beingupdated and, hence, slightly different values were used in thisstudy and are reported in Table 5.

Alkyl Radical Decomposition. The rate constants for alkylradical decomposition were estimated in the reverse, exothermicdirection, that is, the addition of an alkyl radical to an alkene.These expressions are taken from a study by Curran65 for C1-C4 alkyl and alkoxyl radical decompositions, Table 6.

The decomposition of 2-methylhex-1-en-6-yl (aC˙ 7H13-f) radicalto form ethylene and 2-methylbut-1-en-4-yl (aC˙ 5H9-d) radical,Figure 14, was estimated as the addition of aC˙ 5H9-d radical toethylene with a rate constant expression of 1.32× 104T2.48exp-(-6130 cal mol-1/RT) cm3 mol-1 s-1, equivalent to that forethyl radical addition to a carbon atom in ethylene as recom-mended by Curran.65 Furthermore, these rate constants wereused for all reactions in which ethyl or a larger radical adds toethylene, to produce an alkenyl radical.

In the same way, the rate constant of 8.80× 103T2.48 exp-(-6130 cal mol-1/RT) cm3 mol-1 s-1, proposed by Curran forethyl radical addition to an external doubly bound carbon inpropene producing an alkenyl radical, was used for all additions

TABLE 3: Continued


273. 1,2,5C6H8 + CH3 ) a,eC7H11-c 1.89E+03 2.7 6850 286. C2H4 + C2H4 ) C4H8-1,4 6.63E+14 -0.6 0274. 1,3C4H6 + CH3 ) 2C5H9-1 1.76E+04 2.5 6130 287. C2H4 + C3H6 ) C5H10-1,4 6.63E+14 -0.6 0275. 1,3C4H6 + CH3 ) 1C5H9-3 1.76E+04 2.5 6130 288. aC3H5 + C4H91 ) 1C7H14 6.63E+14 -0.6 0276. C2H3 + aC3H4 ) iC5H7 9.45E+02 2.7 6850 289. 1C4H7-4 + nC3H7 ) 1C7H14 6.63E+14 -0.6 0277. 1,3C5H8 + C2H5 ) kC7H13-j 9.45E+02 2.7 6850 290. 1C5H9-5 + C2H5 ) 1C7H14 6.63E+14 -0.6 0278. aC3H4 + 1C4H7-4 ) a,eC7H11-a2 9.45E+02 2.7 6850 291. nC4H5 + CH3 ) 1,3C5H8 6.63E+14 -0.6 0279. C4H8-1,4+ C3H6 ) C7H14-a,f 9.45E+02 2.7 6850 292. CYCHEXENE) 1,3C4H6 + C2H4 3.64E+62 -13.8 92030280. aC3H5 + aC3H5 ) 1,5C6H10 6.63E+14 -0.6 0 293. 1,3CYCHEX) C6H6 + H2 2.36E+19 -1.4 67530281. aC3H5 + 1C4H7-4 ) 1,6C7H12 6.63E+14 -0.6 0 294. 1,3CYCHEX) 1,3CYCHEX-5+ H 5.01E+15 0.0 72600282. aC3H5 + iC4H7 ) 1,5C7H12 6.63E+14 -0.6 0 295. 1,3CYCHEX-5) C6H6 + H 4.80E+25 -3.5 33480283. iC4H7 + aC3H5 ) g,kC7H12 6.63E+14 -0.6 0 296. 1,3CYCHEX-5) nC4H5 + C2H2 2.51E+14 0.7 41827284. 1,5C6H9-3 + CH3 ) g,kC7H12 6.63E+14 -0.6 0 297. 2C6H10-1,6) 1,3C4H6 + C2H4 2.07E+13 0.4 38219285. iC4H5 + CH3 ) iC5H8 6.63E+14 -0.6 0 298. 3C6H10-1,6) C2H4 + C2H2 + C2H4 2.07E+13 0.4 38219

TABLE 4: MCH Decomposition Forming Diradicals;s/cal/mol Units

radical type A n EA

MCH f C7H14-1,6 2.5E+16 0.0 82457MCH f C7H14-a,f 2.5E+16 0.0 86042MCH f C7H14-g,l 2.5E+16 0.0 86042

TABLE 5: Arrhenius Parameters per H-Atom; cm 3/mol/s/cal Units

abstraction by C-type A n EA

1° 2.22E+05 2.54 6756H 2° 6.50E+05 2.40 4471

3° 6.02E+05 2.40 25831° 1.76E+09 0.97 1586

OH 2° 2.34E+07 1.61 -353° 5.73E+10 0.51 631° 1.51E-01 3.65 7154

CH3 2° 7.55E-01 3.46 54813° 6.01E-10 6.36 8931° 7.93E+03 2.55 16 494

HO2 2° 4.82E+03 2.60 13 9103° 3.61E+03 2.55 10 5321° 7.93E+03 2.55 16 494

CH3O2 2° 4.82E+03 2.60 13 9103° 3.61E+03 2.55 10 5321° 7.23E+10 0.00 6458

CH3O 2° 7.25E+10 0.00 45713° 2.29E+10 0.00 28731° 3.27E+05 2.43 4750

O 2° 2.76E+05 2.45 28303° 3.83E+05 2.41 11401° 1.00E+13 0.00 52 290

O2 2° 1.00E+13 0.00 49 6403° 1.00E+13 0.00 48 200

TABLE 6: C4 Radical Addition; cm3/mol/s/cal Units

radical type A n EA

External CRadical AdditionCH3 + 1,3C4H6 1.76E+04 2.48 6130C2H5 (or larger)+ C2H4 1.32E+04 2.48 6130C2H5 (or larger)+ C3H6 (or larger) 8.80E+03 2.48 6130

Internal CRadical AdditionCH3 + C3H6 (or larger) 1.89E+03 2.67 6850C2H5 (or larger)+ C3H6 (or larger) 9.45E+02 2.67 6850

Figure 6. Ignition delay times as a function of inverse temperature at0.099% MCH/Ar,φ ) 0.105, andp5 ) 1.02 atm: (9) this study; (O)Hawthorn and Nixon.40


of ethyl or larger radicals, to the external doubly bonded carbonin propene or larger alkenes, for example:

C2H5 adding to isoprene (iC5H8) to form 3-methylhex-1-en-3-yl (gC7H13-i), Figure 16. The rate constant of 1.76× 104T2.48

exp(-6130 cal mol-1/RT) cm3 mol-1 s-1, recommended byCurran for methyl addition to ethylene, was used for methyladdition to a terminal carbon atom in buta-1,3-diene, forexample, methyl addition to buta-1,3-diene forming the pent-1-en-3-yl (1C5H9-3) radical.

With the same assumptions, the rate constant expression of1.89 × 103T2.67 exp(-6850 cal mol-1/RT) cm3 mol-1 s-1

recommended by Curran for methyl addition to the internaldoubly bound carbon in propene was also used for methyladdition to an equivalent carbon in larger radicals, for example,methyl addition to cyclohexene forming 1-methylcyclohex-2-yl (MCH-R2) radical, Figure 16:

Similarly, the rate constant of 9.45× 102T2.67 exp(-6850 calmol-1/RT) cm3 mol-1 s-1, for ethyl radical addition to theinternal doubly bound carbon atom in propene, was used in all

Figure 7. Ignition delay times for a 1.0% MCH in Ar mixture,φ )1.0 at (9) 1.0 atm, (O) 2.0 atm, and (2) 4.0 atm. Symbols areexperimental results; lines are model predictions. The dashed linecorresponds to open symbols.

Figure 8. Ignition delay times for MCH in Ar mixtures at 1 atm: (9)φ ) 2.0 (1.0% MCH, 5.25% O2); (O) φ ) 1.0 (1.0% MCH, 10.5%O2); (2) φ ) 0.5 (0.5% MCH, 10.5% O2). Symbols are experimentalresults; lines are model predictions.

Figure 9. MCH profiles versus time in a flow reactor near-pyrolysis39

at 1664 ppm MCH, 180 ppm O2, and 1 atm: (9) 1058 K; (O) 1108 K;(2) 1154 K; (]) 1192 K. Symbols are experimental results, lines aremodel predictions.

Figure 10. Species profiles versus time in a flow reactor near-pyrolysis39 at 1664 ppm MCH, 180 ppm O2, 1155 K, and 1 atm: (9)MCH; (b) C2H4; (4) 1,3C4H6; (1) CH4; (]) C3H6; (f) iC5H8. Symbolsare experimental results; lines are model predictions. The dashed linecorresponds to open symbols. Time shift) -0.005 s.

Figure 11. Species profiles versus time in a flow reactor oxidation39

at 1815 ppm MCH,φ ) 1.3, 1160 K, and 1 atm: (9) MCH; (b) C2H4;(4) 1,3C4H6; (1) CH4; (]) C3H6; (f) iC5H8. Symbols are experimentalresults; lines are model predictions. The dashed line corresponds toopen symbols. The dotted line corresponds to penta-1,3-diene simula-tion, for which no experimental data were available.


instances where ethyl or a larger radical added to an internaldoubly bound carbon atom in an alkene, for example:

ethyl addition to penta-1,3-diene (1,3C5H8) forming 4-methyl-hex-1-en-3-yl (kC7H13-j) radical, Figure 17.

Alkyl Radical Isomerization. The rate constants used forradical isomerization in this study are consistent with thoserecommended by Matheu et al.,66 who presented rate constants

for these reactions derived from the results of high level(B3LYP-ccpVDZ) quantum calculations by Sumanthi. Thefrequency (A-factor) and temperature dependent (Tn) termsassociated with all 1,2 H-shifts are equivalent, with the activationenergy of each reaction being dependent on the type (primary,secondary, tertiary, or allylic) of both the initial and final positionof the H atom undergoing isomerization. Frequency factors andtemperature-dependent terms are also identical for all 1,3H-shifts and for all 1,4 H-shifts, with the exception of 1,4primary radical isomerization forming an allylic radical whichhas unique values. These values have been adopted in the currentstudy and are provided in Table 7.11

Matheu defines a “pf s” H-shift as a H-shift within aprimary alkyl radical leading to the formation of a secondaryalkyl radical and uses the same principle when naming all otherH-shifts. However, in this study the type of H atom was usedto describe the H-shift. Thus, a H-shift in which the H atom

Figure 12. Species profiles versus time in a flow reactor oxidation39

at 1815 ppm MCH,φ ) 1.3, 1160 K, and 1 atm; (9) CO × 0.3; (b)CO2 × 0.6. Lines: solid) CO× 0.3 and CO2 × 0.6, dash) H2, dashdot ) H2O × 0.3, dot ) CH2O. Symbols are experimental results;lines are model predictions.

Figure 13. Flux analysis for MCH consumption at 1815 ppm MCH,φ ) 1.3, 1160 K, and 1 atm, at 10% consumption.

Figure 14. MCH-R1 decomposition. Conditions as per Figure 13.





moves from a primary carbon atom to a secondary carbon atomis defined as a “pf s” H-shift.

Further analysis of the rate constants recommended byMatheu et al.66 indicated that there were general trends inactivation energy for the isomerization of primary, secondary,tertiary, and allylic H atoms. The trends observed for 1,2, 1,3,

and 1,4 H-shifts were used to extrapolate Matheu’s work andestimate activation energies for isomerization reactions thatoccurred during MCH oxidation but were not previouslyspecified. Table 8 shows the calculatedEA values (in bold),together with theEA values that were defined by Matheu forrelevant reactions. A complete list of rate constants used forisomerization reactions is provided in Table 7.

The rate expression of 3.56× 1010T0.88 exp(-31500 calmol-1/RT) cm3 mol-1 s-1 for a 1,2 allylic sf s H-shift, forexample,

pent-1-en-4-yl (1C˙ 5H9-4) forming pent-1-en-3-yl (1C˙ 5H9-3), wastaken directly from Matheu and used in this study, Table 7.

The rate expression of 3.67× 1012T-0.6 exp(-15300 calmol-1/RT) cm3 mol-1 s-1 was used by Matheu for 1,4 H-shiftsin which a primary radical isomerized to form a secondary allylicradical. In this study theseA andTn terms were used for all 1,4H-shifts, Table 7. AnEA of 15.3 kcal mol-1 was used for 1,4allylic s f p H-shifts, for example:

2-methylhex-1-en-6-yl (aC˙ 7H13-f) radical forming 2-methylhex-1-en-3-yl (aC˙ 7H13-c) radical, Figure 14.

The activation energy was decreased by 2.3 kcal mol-1 to13.0 kcal mol-1 for 1,4 allylic t f p H-shifts), Table 7, forexample:

3-methylhex-1-en-6-yl (gC˙ 7H13-l) radical forming 3-methylhex-1-en-3-yl (gC7H13-i) radical, Figure 16. This corresponds to thedifference in activation energy between an sf p and a tf pH-shift, Table 8. For 1,4 allylic sf s H-shifts, for example,

hept-1-en-6-yl (1C˙ 7H13-6) radical forming hept-1-en-3-yl(1C7H13-3) radical, Figure 17, the activation energy was insteadincreased by 2.6 kcal mol-1 to 17.9 kcal mol-1, Table 7. Thischange in energy corresponds to the average difference between1,4 sf p and sf s, and tf p and tf s H-shifts, Table 8.

The rate constant expression of 3.67× 1012T-0.6 exp(-15300cal mol-1/RT) cm3 mol-1 s-1 is recommended by Matheu for1,5 H-shifts in which a primary radical isomerizes to form aprimary radical. TheseA-factor andTn values were used in thisstudy for all 1,5 H-shifts that did not result in the formation ofan allylic radical, Table 7. As before theEA value was adjustedfor individual reactions. For instance, when a primary radicalsite on a diradical isomerized through a 1,5 H-shift to asecondary position forming an alkene (that is, sf p H-shifts),for example,

hept-1,6-diyl (C˙ 7H14-1,6) forming hept-2-ene (2C7H14), theactivation energy was decreased by 2.5 kcal mol-1 to 12.8 kcalmol-1, which corresponds to the difference in energy betweena p f p and sf p 1,4 H-shifts, Table 8.

In the case of a primary radical site on a diradical isomerizingto a tertiary site forming an alkene (that is, a tf p 1,5 H-shift),for example,

TABLE 7: Isomerization Reactions; s/cal/mol Units

A n EA

1,2 H-Shiftsallylic s f s 3.56E+10 0.88 31 500

1,4 H-Shiftsallylic s f p 3.67E+12 -0.67 15 300allylic t f p 3.67E+12 -0.67 13 000allylic s f s 3.67E+12 -0.67 17 900

1,5 H-Shiftss f p 3.67E+12 -0.67 12 800t f p 3.67E+12 -0.67 10 500s f s 3.67E+12 -0.67 15 300

1,5 H-shifts Forming Allylic Radicalsallylic sa f p 2.80E+10 0.00 7 500allylic pb f p 3.67E+12 -0.60 10 000

1,7 H-Shiftshept-2-ene-7-ylf hept-2-ene-1-yl 2.80E+10 0.00 31 640hept-1,6-diylf hept-1-ene 2.80E+10 0.00 25 800

a An allylic structure is created outside the ring structure in thetransition state.b A double bond is created inside the ring structure inthe transition state.

TABLE 8: EA (cal mol-1) Calculations for 1,4 and 1,5H-Shiftsa

isomerization type of H-atom EA ∆EA

1,4 H-Shifts1,4 sf p 20 6001,4 tf p 18 300 -23001,4 allylic sf p 15 3001,4 allylic t f p 13 000 -23001,4 sf p 20 6001,4 sf s 23 500 +29001,4 tf p 18 3001,4 tf s 20 600 +23001,4 allylic sf p 15 3001,4 allylic sf s 17 900 +2600b

1,5 H-Shifts1,4 pf p 23 1001,4 sf p 20 600 -25001,5 pf p 15 3001,5 sf p 12 800 -25001,4 pf p 23 1001,4 tf p 18 300 -48001,5 pf p 15 3001,5 t f p 10 500 -48001,3 pf p 38 8001,3 sf s 38 200 -6001,4 pf p 23 1001,4 sf s 23 500 +4001,5 pf p 15 3001,5 sf s 15 300 +0c

1,5 H-Shifts Forming Allylic Radicals1,4 pf p 23 1001,4 allylic sf p 15 300 -78001,5 pf p 15 3001,5 allylic sf p 7500 -78001,4 sf p 20 6001,4 allylic sf p 15 300 -53001,5 pf p 15 3001,5 allylic p f p 10 000 -5300

a Bold entries haveEA values derived from those recommended byMatheu.66 b +2600) average∆EA for two previous entries, that is,+2900 and 2300.c 0 ≈ average∆EA for two previous entries.


2-methylhex-1,6-diyl (C˙ 7H14-a,f) forming 2-methylhex-1-ene(aC7H14), the activation energy was decreased by 4.8 kcal mol-1

to 10.5 kcal mol-1, which is equal to the difference in energybetween 1,4 pf p and tf p H-shifts, Table 8.

For reactions in which a secondary radical site on a diradicalisomerized by a 1,5 H-shift to a secondary site forming an alkene(that is, a sf s H-shift), for example,

hept-1,6-diyl (C˙ 7H14-1,6) forming hept-1-ene (1C7H14), nochanges were made to the activation energy due to the fact thatthe rate expressions recommended by Matheu have very similaractivation energies for both 1,3 and 1,4 pf p and sf sH-shifts, Table 8.

In cases where allylic radicals were formed from 1,5 H-shiftsdifferent rules were used to estimate rate constants for thereactions. When 1,5 allylic sf p H-shifts involved theformation of an allylic radical outside the ring structure in thetransition state, for example,

hept-1-en-7-yl (1C˙ 7H13-7) forming hept-1-en-3-yl (1C˙ 7H13-3),the rate expression was estimated using theA-factor andTn valuerecommended by Matheu for 1,6 H-shifts, Table 7. Thiscompensates for the loss of an extra internal rotor in themolecule as an allylic radical is formed in a positionâ to the1,5 shift. The activation energy was taken as 7.50 kcal mol-1,which is the activation energy for pf p 1,5 H-shifts (15.3 kcalmol-1) minus the difference in energy between 1,4 pf p andallylic s f p (7.8 kcal mol-1), Table 8. This yielded a rateconstant of 2.8× 1010 exp(-7500/RT) cm3 mol-1 s-1, Table 7.

For 1,5 allyl pf p H-shifts, where the double bond is presentin the ring structure of the transition state, that is, in the case of

pent-2-en-5-yl (2C˙ 5H9-5) forming pent-2-en-1-yl (2C˙ 5H9-1), theA-factor andTn value recommended by Matheu for 1,4 H-shiftsforming a secondary allylic from a primary radical were used,due to one less rotor being lost in the transition state, Table 7.The activation energy for 1,5 pf p H-shifts was decreased by5.3 kcal mol-1, which corresponds to the difference in activationenergy between sf p (20.6 kcal mol-1) and allylic sf p (15.3kcal mol-1) 1,4 H-shifts, Table 8. This resulted in a rate constantof 3.67× 1012T-0.6 exp(-10000 cal mol-1/RT) cm3 mol-1 s-1,Table 7.

In the case of

hept-2-en-7-yl (2C˙ 7H13-7) f hept-2-en-1-yl (2C˙ 7H13-1), a 1,7H-shift occurred and a double bond was present in the seven-membered cyclic transition state. For this isomerization theA-factor andTn value recommended by Matheu for 1,6 H-shiftswere used due to one less rotor being lost in transition stateformation, Table 7. The activation energy was estimated usingthe Evans-Polanyi rule devised by Matheu with the ring strainbeing estimated as 11.0 kcal mol-1 (10.0 kcal mol-1 for an eight-membered cyclic transition state67 plus 1.0 kcal mol-1 to accountfor the extra strain caused by the double bond in the cyclic

transition state68). This leads to a rate expression of 2.8× 1010

exp(-31640 cal mol-1/RT) cm3 mol-1 s-1, Table 7.For 1,7 H-shifts from a primary to primary position, which

resulted in a double bond being formed inside the ring structureof the transition state, that is

hept-1,6-diyl (C˙ 7H14-1,6) forming hept-1-ene (1C7H14), theA-factor andTn value recommended by Matheu for 1,6 H-shiftswere used. The activation energy was estimated to be 25.8 kcalmol-1. This value was calculated by increasing the ring strainof cyclooctane by the average difference between the ring strainof four-, five-, and six-membered cyclic hydrocarbons and theactivation energy for primary to primary H-shifts that proceedthrough the corresponding four-, five-, and six-memberedtransition states, Table 11. As before, 1.0 kcal mol-1 was addedto compensate for the extra ring strain incurred by the formationof a double bond inside the ring structure of the transition state.This resulted in a rate constant of 2.8× 1010 exp(-25800 calmol-1/RT) cm3 mol-1 s-1, Table 7.

Other Alkene/Alkenyl Radical Decompositions.The rateconstant for cyclohexa-1,3-diene decomposition forming ben-zene and molecular hydrogen was estimated to be 2.36×1019T-1.38exp(-67530 cal mol-1/RT) cm3 mol-1 s-1 by creatinga best fit though the data provided for this reaction by Orchardet al.69 and Alfassi et al.70 A rate constant of 5.01× 1015 exp-(-72600 cal mol-1/RT) cm3 mol-1 s-1 was reported by Dean71

for cyclohexa-1,3-diene decomposition forming cyclohexa-2,4-dien-1-yl and a hydrogen atom and was used in this study.

The cyclohexa-2,4-dien-1-yl radical (1,3CYCHEX-5) candecompose via two pathways:

For the first reaction, which produces C2H2 and nC4H5 radicals,the rate constant of 2.51× 1014T0.7 exp(-41830 cal mol-1/RT)cm3 mol-1 s-1 recommended by Weissman et al.72 was used.The rate constant of 4.80× 1025T-3.5 exp(-33480 cal mol-1/RT) cm3 mol-1 s-1 for the decomposition producing benzeneand a hydrogen atom was estimated by generating a best fit ofthe data provided by Dean71 and Nicovich et al.73 for thisreaction.

In the same way the rate constant for cyclohexene decom-position to give buta-1,3-diene and ethylene was estimated tobe 3.64× 1062T-13.8exp(-92030 cal mol-1/RT) cm3 mol-1 s-1

by creating a best fit of the data provided for this reaction usingthe studies of Lewis et al.,74 Hidaka et al.,75 Skinner et al.,76

Newman et al.,77 Barnard et al.,78 and Tsang.79,80

Ring Opening Reactions.The ring opening reactions werewritten in the reverse intramolecular alkenyl radical ring closuredirection and the rate constants were estimated in this direction.Rate constants for ring closure leading to the formation of cyclicradical species, for example:

2-methylhex-1-en-6-yl (aC˙ 7H13-f) forming 1-methylcyclohex-1-yl (MCH-R1), Figure 14, were derived from those reportedby Matheu et al.,66 who presented high-pressure-limit rate rulesfor radical additions and ring opening/closing reactions. Radicaladdition rules developed by Matheu et al.66 were taken from


the work of Curran and co-workers,54 and in the case of vinyladdition, from Fahr and Stein.81 Rate rules for cycloalkyl ringclosures or openings were taken by Matheu et al. from modelreactions listed by Newcomb where available;82 otherwise, theywere estimated using both simple (rigid-rotor harmonic oscil-lator) transition-state theory (TST) and detailed ab initiocalculations. There are two distinct types of ring closure definedby Matheu et al. as “exo” and “endo” ring opening/closure; for“exo” ring closure the radical site isâ to the carbon ring (radicaladdition in the alkenyl radical occurs on the internal doublybound carbon), and for “endo” ring closure the radical site ison the ring itself (radical addition occurs on the terminal doublybound carbon).

In this study “endo” 1,6 intramolecular ring closures havebeen treated by taking the rate constant expression of Matheuet al. for primary carbon radical addition to an external doublybound carbon atom, that is, 1.0× 108T0.86exp(-5900 cal mol-1/RT) cm3 mol-1 s-1. Though theseA-factor andTn values wereused in this study for all “endo” 1,6 intramolecular ring closures,the activation energy,EA, was altered for individual reactionsto allow for (i) increased ring strain, (ii) different radical types,and (iii) the location of the carbon atom to which additionoccurred. Alterations to theEA value were made by analyzingthe differences betweenEA values presented by Matheu et al.for relevant reactions and changing theEA value of 5.9 kcalmol-1 by the same amount. More detailed explanations are givenin the following section. Table 9 shows the calculatedEA values(in bold), along with theEA values that were defined by Matheuet al. for relevant reactions, with Table 10 listing theA, Tn, andEA values for each type of ring closing reaction.

The rate constant of 1.0× 108T0.86 exp(-5900 cal mol-1/RT) cm3 mol-1 s-1 was used without modification in caseswhere addition occurred specifically as outlined by Matheu etal., for example, 2-methylhex-1-en-6-yl (aC7H13-f) forming1-methylcyclohex-1-yl (MCH-R1) radical, Figure 14 (any effectscaused by the methyl group were ignored), Table 10.

For “endo” 1,6 primary carbon radical addition to an internaldoubly bonded carbon, for example,

hept-2-en-7-yl (2C7H13-7) forming 1-methylcyclohex-2-yl (MCH-R2), Figure 15, the activation energy for addition to a terminaldoubly bound carbon was increased by 2.80 kcal mol-1 to 8.70kcal mol-1, consistent with the increase in activation energybetween alkyl radical addition to the terminal and internal carbonatoms of a double bond as reported by Matheu et al.,66 Tables9 and 10.

In the case of “endo” 1,6 secondary carbon radical additionto an external doubly bound carbon, for example,

hept-1-en-6-yl (1C˙ 7H13-6) forming 1-methylcyclohex-3-yl (MCH-

TABLE 9: EA (cal mol-1) Calculations for Ring Closing Reactionsa

EA ∆EA

Intra 1,6 Addition (Endo)alkyl addn to terminal C atom of double bond 7800alkyl addn to internal C atom of double bond 10 600 +2800O 1° C radical addition to external CdC 5900O 1° C4 radical addition to internal C dC 8700 +28001,4 sf p H-shift 20 6001,4 sf s H-shift 23 500 +29001,4 tf p H-shift 18 3001,4 tf s H-shift 20 600 +2300O 1° C radical addition to external CdC 5900O 2° C4 radical addition to external CdC 8500 +2600b

Intra 1,6 Addition (Exo)O cyclo-propyl (endo) ring opening 21 900O cyclo-propyl (exo) ring opening 7060 divided by 3.1O 1° C radical addition to external CdC (endo) 5900O 2° C radical addition to external CdC (exo) 1903 divided by 3.1

a Bold entries haveEA values derived from those recommended by Matheu;66 all others are taken directly from Matheu.b +2600) average∆EA

for two previous entries, that is,+2900 and+2300.c O denotes ring closing/opening reactions.

TABLE 10: Ring Closing Reactions; s/cal/mol Units

A n EA

Intra 1,6 Addition (Endo)1° C radical addition to external CdC 1.0E+08 0.86 59001° C radical addition to internal CdC 1.0E+08 0.86 87002° C radical addition to external CdC 1.0E+08 0.86 85001° C radical addition to external CdCa 1.0E+08 0.86 6900

Intra 1,6 Addition (Exo)1° C radical addition to internal CdC 1.0E+08 0.86 1903

a A CdC bond is contained in the ring structure of the transitionstate.

TABLE 11: Differences between Ring Strain andEA (calmol-1) for a H-Shift Incorporating an Equivalent TransitionStatea

ring strain H-shiftEA for

p f p H-shift ∆EA

cyclobutane 26 300 1,3 38 800 12 500cyclopentane 6500 1,4 23 100 16 600cyclohexane 0 1,5 15 300 15 300cycloheptene 10 000 1,7 24 800 14 800b

a Bold entries are calculated values.b 14 800) average differencebetweenEA and ring strain.


R3) radical, the activation energy was increased by 2.60 kcalmol-1 to 8.50 kcal mol-1, which is the average difference inactivation energy between 1,4 sf p and sf s, and tf p andt f s H-shifts, Tables 9 and 10.

For “endo” 1,6 primary carbon radical addition to an externaldoubly bonded carbon atom when an extra double bond waspresent in the ring, for example,

hexa-1,3-dien-6-yl (1,3C˙ 6H9-6) forming cyclohex-1-en-3-yl(cycC6H9-3) radical, 1.0 kcal mol-1 was added to the activationenergy and a value of 6.9 kcal mol-1 was used, Table 10. Thiscorresponds to the difference in ring strain between cyclohexaneand cyclohexene68 and compensates for the extra energy neededfor the ring structure to be formed.

For “exo” 1,6 intramolecular cyclo-alkyl ring closing, thatis, where the radical site isâ to the carbon ring, for example,

hept-1-en-7-yl (1C˙ 7H13-7) forming CYCHEXCH2, the same rateconstant of 1.0× 108T0.86exp(-5900 cal mol-1/RT) cm3 mol-1

s-1 was used but the activation energy was reduced by a factorof 3.1, yielding a value of 1903 cal mol-1, Tables 9 and 10.This corresponds to the ratio of the activation energies for“endo” and “exo” cyclopropyl ring opening. Although this valueis perhaps not entirely suitable, the sensitivity analysis does nothighlight the rate of this reaction as being very significant forMCH combustion.

Results and Discussion

Shock Tube Ignition Delays.Shock tube ignition delay timesfor MCH/O2/Ar mixtures were measured behind reflected shockwaves over the temperature range 1250-2100 K at reflectedshock pressures of 1.0, 2.0, and 4.0 atm, and equivalence ratios,φ, of 0.5, 1.0, and 2.0, Figures 6-8. Delay times of over 1 msto fewer than 50µs were recorded. Shorter delay times wereobserved, but the corresponding percentage error accounted insome cases for a large portion of the delay. For this reason delaytimes of less than 50µs were ignored.

To perform a comparative study, the shock tube experimentsof Hawthorn and Nixon40 were repeated. Although theseexperiments had been performed behind incident shock waves,their conditions of temperature and pressure were reproducedin the current study to compare the results directly. Figure 6depicts ignition delay times as a function of inverse temperatureat 0.099% MCH/Ar,φ ) 0.105, at a pressure of 1.02 atmmeasured behind the reflected shock in the present study andbehind the incident shock by Hawthorn and Nixon. For theseexperiments, the ignition delay time was defined as the timebetween the shock arrival at the end wall and the first lightemission, as measured from the end wall. For experiments inthis study, the same end-on detection diagnostic was used (withthe filter removed). The data generated in the current study andthose recorded by Hawthorn and Nixon are in very goodagreement, although it could be argued that the data of Hawthornand Nixon shows a slightly higher activation energy relative tothat measured in the current study.

The ignition delay times obtained for a 1.0% MCH/10.5%O2 stoichiometric (φ ) 1.0) mixture, over the temperature range1300-2200 K, at pressures of 1.0, 2.0, and 4.0 atm, are shownin Figure 7, together with model predictions. The ignition delaytimes decrease as the reflected shock pressure is increasedprogressively from 1.0 to 2.0 and then 4.0 atm, in agreementwith most studies carried out to date45 on hydrocarbons. Theignition delay times simulated using the MCH mechanism, inconjunction with the Chemkin III suite of programs, are slightlyfaster than those measured experimentally but they accuratelycapture the pressure dependence.

Ignition delay times for different MCH/O2 mixtures measuredat atmospheric pressure are shown in Figure 8, together withsimulated ignition delay times. Increasing the oxygen concentra-tion from 5.25% (φ ) 2.0) to 10.50% (φ ) 1.0), with a constantMCH concentration of 1.0%, led to a significant reduction indelay times. This negative power dependence of oxygen is inaccordance with previous work.45 Conversely, as the MCHconcentration was increased from 0.5% (φ ) 0.5) to 1.0% (φ) 1.0), with the O2 concentration constant at 10.5%, the ignitiondelay times increased but the magnitude of this effect was muchless pronounced than in the case of oxygen. This observationis also in agreement with previous studies of hydrocarbons. Theconcentration of argon in all three mixtures was essentiallyconstant and so did not influence the delay times. Model-predicted ignition delay times are slightly faster than thosemeasured but do accurately capture the influence of fuel andoxygen concentration.

Modeling of Turbulent Flow Reactor. The MCH mecha-nism was used to simulate the species profiles measured forMCH near-pyrolysis and oxidation in a turbulent flow reactorby Zeppieri et al.39 MCH pyrolysis decay profiles weredetermined as functions of time at 1664 ppm MCH, 180 ppmO2, 1 atm, and at four temperatures (1058, 1108, 1154, and 1192K). The detailed chemical kinetic mechanism was used tosimulate these data and a comparison of the experiment andmodel predictions are provided in Figure 9. Overall, the modelcaptures the experimental trends of fuel consumption in that,as the reactor temperature increases, the fuel decay timedecreases and the profiles become much steeper.

The experiments also produced species profiles versus timefor MCH near-pyrolysis at 1664 ppm MCH, 180 ppm O2, 1155K, 1 atm, Figure 10. In addition, species profiles were alsorecorded for MCH oxidation at 1815 ppm MCH,φ ) 1.3, 1160K, 1 atm, Figure 11. Major products, for both oxidation andpyrolysis, included methane (CH4), ethylene (C2H4), propene(C3H6), buta-1,3-diene (1,3C4H6), and isoprene (iC5H8). Aro-matic and cyclic hydrocarbon species were among the minorproducts recorded, as were ethane and some unsaturated C2,C3, and C4 hydrocarbons. The detailed chemical kinetic mech-anism created in this study was used to simulate both thepyrolysis and oxidation experiments, Figures 10 and 11.

All simulated species profiles for the pyrolysis experimentswere time-shifted by-5 ms relative to the experiments to allowfor the experimental uncertainties of the time-temperaturehistory of the in-flowing reactants. Time-shifting the simulatedpyrolysis data by this amount led to the best overall agreementbetween the experimental and predicted fuel-consumptionprofiles, Figure 10. The MCH profile does not decrease asrapidly as the experimental data between 5 and 25 ms, but theoverall agreement is quite good. In general, the shape andmagnitude of the major species profiles are in good agreementwith the experimental data but methane production is under-predicted by≈20%.


In the oxidative case, for which no time-shifting wasnecessary, the MCH profile does not fall off as rapidly as theexperimental data between 0 and 25 ms, Figure 11. Again, mostof the major species profiles are in generally good agreement.Species production does not occur as rapidly as the experimentaldata suggest between 0 and 5 ms, and the model also predictsthat all species profiles will have dropped to near zero at≈58ms whereas the experimental data drop to near zero at≈63ms. Again, methane is underpredicted by≈20%.

Some species predicted by the model were not observed underthe oxidation experimental conditions. Hydrogen, water, andformaldehyde were predicted in relatively large quantities, andpenta-1,3-diene was also predicted, albeit on a much smallerscale, Figures 11 and 12. The analytical techniques (MS andGC-MS) employed may not have been entirely suitable for thedetection and quantification of hydrogen, CH2O, and H2O.83

Rate of Production Analysis.A rate of production analysiswas performed using Chemkin in the case of the MCH oxidationin the flow reactor, at 1815 ppm MCH, 14 660 ppm O2 (φ )1.3), 1160 K, 1 atm, at a time corresponding to 10% MCHconsumption. This analysis revealed that oxidation occurs byH-atom abstraction from the fuel, Figure 13. Hydrogen abstrac-tion by molecular oxygen, H˙ and Oatoms, and C˙ H3, HO2, OH,and CH3O radicals, occurs at primary, secondary, and tertiary(1°, 2°, and 3°) sites, leading to the formation of five differentmethylcyclohexyl radicals, Figure 13. The rates for abstractionby different radical species were assumed to be equivalent toH-atom abstraction rates from similar sites in other alkanemolecules and the rate constants used are shown in Table 5.

Figure 13 shows the relative importance of the abstractionpathways. Flux analysis shows that secondary H-atom abstrac-tion dominates, leading preferentially to the formation of MCH-R2, MCH-R3, and MCH-R4. MCH-R2 and MCH-R3 areformed in equal concentrations (≈27%) whereas MCH-R4accounts for approximately 13% of the fuel. This is due to thedegeneracy of the system, as there are four equivalent hydrogenatoms available to form MCH-R2 and MCH-R3 but there areonly two hydrogen atoms in the case of MCH-R4. Tertiaryhydrogen atom abstraction leads to the formation of MCH-R1(≈14%) whereas primary hydrogen atom abstraction producesthe cyclohexyl-methylene radical (≈6%). Finally, the unimo-lecular decomposition of MCH producing a cyclohexyl and amethyl radical accounts for approximately 11% of MCH.

The decomposition of the methylcylohexyl radicals MCH-R1, MCH-R3, and MCH-R4 are detailed in Figures 14, 17, and18, respectively. The two ring opening steps and theâ-scissionstep, by which MCH-R2 decomposes are illustrated in Figures15 and 16. MCH-R1, which accounts for 14% of the fuel,undergoes a ring opening reaction to form 2-methylhex-1-en-6-yl (aC7H13-f), Figure 14. This in turn undergoesâ-scission(90%) forming 2-methylbut-1-en-4-yl (aC˙ 5H9-d) and ethylene.The aC5H9-d forms tC3H5 plus ethylene byâ-scission (73%)and isoprene (iC˙ 5H8) plus H-atom (27%). The other 10% of

aC7H13-f isomerizes to form 2-methylhex-1-en-3-yl (aC˙ 7H13-c), which then forms isoprene plus C˙ 2H5.

MCH-R2 decomposes via three different pathways, Figures15 and 16:

1. 61% forms hept-2-en-7-yl (2C˙ 7H13-7), which proceedsvia â-scission (69%) to yield pent-2-en-5-yl (2C˙ 5H9-5) plusethylene. 2C˙ 5H9-5 then isomerizes to yield pent-2-en-1-yl(2C5H9-1) (95%), which in turn forms butadiene plusmethyl radical (97%). Another 29% of 2C˙ 7H13-7 isomerizesto form hept-2-en-4-yl (2C˙ 7H13-4), which then yields penta-1,3-diene (1,3C5H8) plus C2H5 radical, Figure 15.2. 16% of MCH-R2 undergoesâ-scission, producingcyclohexene plus methyl radical, Figure 16.3. 22% undergoes ring opening, producing 3-methylhex-1-en-6-yl (gC7H13-l). gC7H13-l then decomposes via twodifferent pathways. 37% undergoes isomerization to form3-methylhex-1-en-3-yl (gC˙ 7H13-i), which then yields iso-prene plus C˙ 2H5. The other 63% of gC˙ 7H13-l undergoesâ-scission to give 2-methylbut-3-en-1-yl (cC˙ 5H9-a) plusethylene. cC˙ 5H9-a yields propene plus C˙ 2H3 (69%) andbutadiene plus methyl radical (30%), Figure 16.The decomposition of MCH-R3 occurs via two major ring

opening pathways, Figure 17. In the first pathway, 60% forms3-methylhex-5-en-1-yl (kC˙ 7H13-g). This primarily (93%) un-dergoes â-scission to form pent-1-en-4-yl (1C˙ 5H9-4) plusethylene. The 1C˙ 5H9-4 species then either isomerizes (70%) toform pent-1-en-3-yl (1C˙ 5H9-3), which then produces butadieneplus methyl, or forms penta-1,3-diene (1,3C5H8) plus H-atom(29%). In the second pathway, MCH-R3 forms hept-1-en-6-yl(1C7H13-6) (40%) via a different ring opening step. The1C7H13-6 then decomposes viaâ-scission (90%) to form propeneplus but-1-en-4-yl (1C˙ 4H7-4), which in turns yields butadieneplus H-atom (80%). Another 8% of 1C˙ 7H13-6 isomerizes to formhept-1-en-3-yl (1C˙ 7H13-3), which gives butadiene plus nC˙ 3H7.The nC3H7 radical then forms ethylene plus methyl radical.

MCH-R4 decomposes via one ring opening step to form2-methylhex-5-en-1-yl (eC˙ 7H13-a), Figure 18. The majority(85%) of this undergoesâ-scission to form propene plus but-1-en-4-yl (1C4H7-4), which in turn forms butadiene plus H-atom(80%).

Chemical Kinetic Mechanism: Sensitivity Analysis.Theaddition of the species and reactions necessary for MCHoxidation to the original mechanism of Laskin et al. for buta-1,3-diene oxidation, resulted in a chemical kinetic mechanismcontaining 190 species and 904 reversible reactions. To carryout the sensitivity analysis, theA-factor of a reaction wasincreased by a factor of 2 and the altered mechanism was usedto simulate MCH oxidation under particular conditions. Twosensitivity analyses were performed, one showing sensitivityto ignition delay times and the second to fuel consumption inthe flow reactor.

Sensitivity analysis of shock tube ignition delay times wascarried out using a 1% MCH, 10.5% O2 in 88.5% Ar mixture,at a reflected shock pressure of 1.0 atm at temperatures of 1250and 1667 K. For each reaction, the percentage difference inignition delay time for the altered and unaltered mechanismswas measured. A negative sensitivity value indicates a decreasein the ignition delay time and thus an increase in overallreactivity. Figure 19 contains the coefficients for reactionscontained in the Laskin et al.52 buta-1,3-diene and the OÄConaire53 H2/O2 submechanisms. Reactions with a sensitivitycoefficient of less than 3% were excluded. Figure 20 depictssensitivity coefficients for all reactions that were added in this



study for the C5, C6, and C7 submechanisms. In this casereactions with sensitivity coefficients of less than 1% wereexcluded.

The shock tube sensitivity analysis highlighted the largenegative sensitivity of the decomposition reaction of MCH tocyclohexyl and a methyl radical; that is, increasing theA-factorof this reaction caused the ignition delay times to decreasesignificantly at both 1250 and 1667 K. To understand thisobservation a flux analysis was carried out at 1250 K at a timeof 107 µs, that is, the time at which 10% fuel had beenconsumed. This showed that 9% of MCH consumed decomposesto give cyclohexyl and a methyl radical:

The cyclohexyl radical then forms hex-1-en-6-yl (1C6H11-6):

which in turn decomposes to give 90% but-1-en-4-yl (1C˙ 4H7-4)plus ethene (C2H4):

78% of but-1-en-4-yl produces buta-1,3-diene (1,3C4H6) and aH-atom (73%) by C-H bond homolysis, or HO˙ 2 by reactionwith O2 (5%):

The other 21% of the but-1-en-4-yl undergoesâ-scission toproduce ethylene and vinyl (C˙ 2H3) radical:

It should also be noted that, in addition to MCH decomposi-tion forming cyclohexyl and a methyl radical, other reactionsthat favored the formation of C˙ 2H3 also have negative sensitivityvalues, Figure 19. These included

In the same manner the reaction

has a positive sensitivity value because it removes C2H3 and areactive H-atom and replaces them with C2H2 and H2. The factthat this reaction removes a H-atom from the system andcompetes with

contributes to its positive sensitivity coefficient. The latterreaction has a strong negative sensitivity value because itincreases the radical concentration present by reacting H-atomwith O2 to produce two reactive radical species.

The reaction

also has positive sensitivity values due to the fact that itcompetes directly with the pathway by which 1,3C4H6 leads tothe formation of C˙ 2H3 as described above.

For these reasons the chemistry of vinyl radical was scruti-nized. 94% of all C˙ 2H3 present was produced by cyclohexylradical decomposing via 1C˙ 4H7-4 radical as described above.Of the total C2H3 radical formed, 37% reacts with O2 to producetwo HCO molecules via CH2O.

Figure 19. Sensitivity analysis for shock tube ignition delay times at1.0% MCH in Ar mixture,φ ) 1.0, and 1.0 atm: (9) 1250 K; (0)1667 K.

Figure 20. Sensitivity analysis for shock tube ignition delay times at1.0% MCH in Ar mixture,φ ) 1.0, and 1.0 atm; (9) 1250 K; (0)1667 K. (MCHR′ ) all methylcyclohexyl radicals.)

C2H4 + OH f C2H3 + H2O

H2CdCHCH2CH2 (1C4H7-4) f C2H4 + C2H3

aC3H5 + HO2 f OH + C2H3 + CH2O

H2CdCHCHdCH2 (1,3C4H6) + H f C2H4 + C2H3

C2H3 + H f C2H2 + H2

H + O2 f O + OH

H2CdCHCHdCH2 (1,3C4H6) + O f

CH2CHCHCHO+ H

C2H3 + O29837%

HCO + CH2O

CH2O + R9897%

HCO + RH R ) H, O, OH, HO2, CH3


The HCO then reacts

This means that for every molecule of C˙ 2H3 that passes throughthis pathway 1.62 molecules of HO˙ 2 are formed.

HO2 plays an important role in increasing the reactivity ofthe system and its reaction pathways are described later.Therefore the reaction

has a negative sensitivity coefficient value at 1667 K. However,at 1250 K its sensitivity coefficient is positive. This is due tothe fact that at both temperatures two molecules of HC˙ O areformed: one molecule initially as a direct product of the reactionand another as a decomposition product of CH2O. At 1250 Kthe HCO decomposes as shown above with 81% reacting withO2 to form CO and HO˙ 2. This competes with HC˙ O decomposi-tion to form the more reactive H-atom and CO. The analysis at1667 K showed that only 51% of HC˙ O reacted with O2, withanother 45% taking the route that produced a H-atom. Thesefindings have a direct bearing on the reaction of H˙ + O2 f O+ OH. At 1667 K not only is less O2 being consumed but alsomore H-atom is being produced, and vice versa at 1250 K,hence, the negative value at 1667 K and the positive value at1250 K.

The reaction

has a negative sensitivity value at 1250 K. This is due to thefact that at this temperature this reaction competes with the lessreactive combination of two methyl radicals to form ethane. At1667 K the formation of ethane does not occur, hence, the almostnegligible positive value.

The CO produced by the above reactions reacts with eitherOH or HO2 radicals to produce CO2 plus H or OH radicals.The reactions of HO˙ 2 are slightly more complex and interesting.66% of HO2 adds to C˙ H3 to form CH3O + OH radicals. TheCH3O radical then decomposes to CH2O + H-atom (80%) orHO2 (20%). CH2O forms HCO radical, 81% of which yieldsHO2 radical. In this manner the HO˙ 2 concentration is maintainedand HO2 radical continues to react with C˙ H3 to form two reactiveH and OH radicals.

The substantial amount of reactive radicals generated in thisway explains the negative sensitivity values observed for theaddition of HO2 to CH3 producing CH3O and OH radicals. Thepositive sensitivity value of

is due to the fact that it is in direct competition with the morefavorable HO˙ 2 + CH3 pathway. Of the remaining HO˙ 2 7% reactswith aC3H5 to yield OH plus C2H3 and CH2O; 6% reacts withHO2 to produce H2O2 plus O2. The majority of both C˙ 2H3 and

CH2O will react or decompose to give HO˙ 2 with reactive radicalsbeing produced as side products; O2 is also a very reactivespecies. This further explains the negative sensitivity of reactionsthat promote the formation of species that break down to formHO2. For example, the reaction

has negative values because it produces CH2O and by associa-tion HO2.

At both temperatures the reaction

has positive sensitivity coefficients, because two reactive HO˙ 2

radicals are removed from the system.The other major pathway by which the C˙ 2H3 radical decom-

poses is by

This reaction has a high negative sensitivity coefficient and leadsto increased reactivity at both 1250 and 1667 K, because tworadicals are produced from one.

The reaction

shows a positive sensitivity coefficient at 1667 K and a negativesensitivity coefficient at 1250 K, Figure 19. Analysis of thesystem reveals that at 1250 K C˙ H2CHO is consumed at almostthe same rate at which it is formed, with 37% decomposing viathis reaction. Therefore, increasing theA-factor of this reactionhas the direct effect of decreasing the C˙ H2CHO concentrationand producing more CH2CO and the reactive HO˙ 2. At 1667 K,however, C˙ H2CHO is not consumed as readily and, of the C˙ H2-CHO that does react, only 4% takes this pathway, hence, thealmost negligible positive value.

At 1250 K the decomposition reaction

has a positive sensitivity value because it competes with thereaction with O2 that produces CH2CO and HO˙ 2, which has ahigh negative sensitivity coefficient, as described above.

The reactions

all have positive sensitivity values. This can be explained bythe fact that these reactions favor the consumption of reactiveradicals and the formation of more stable species. In addition,reactions in which stables species compete with the fuel inH-atom abstraction reactions by radical species inhibit reactivityand also show positive sensitivity coefficients.

The reaction

HCO + O29881%

CO + HO2

HCO + M 9818%

CO + H + M

C2H3 + O2 f HCO + CH2O

CH2O + CH3 f CH4 + HCO

CH3 + HO29866%

CH3O + OH

CH3O(+ M) 9880%

CH2O + H(+ M)

CH3O + O29819%

CH2O + HO2

CH3 + HO2 985%

CH4 + O2

CH3 + O2 f OH + CH2O

HO2 + HO2 f H2O2 + O2

C2H3 + O29856%

CH2CHO + O

CH2CHO + O29837%

CH2CO + HO2

CH2CHO f CH3 + CO

HO2 + OH f H2O + O2

CH3 + CH3(+ M) f C2H6(+ M)

aC3H5 + H(+ M) f C3H6(+ M)

C3H6 + H f aC3H5 + H2

C3H6 + OH f aC3H5 + H2O

CH2O + OH f HCO + H2O

CH3 + OH f CH2* + H2O


has a negative sensitivity where two radicals combine to formreactive CH2* and H2O.

The sensitivity coefficients for reactions added, in this study,to the original mechanism by Laskin et al. are shown in Figure20. The large negative value for

is due to the fact that it favors the formation of C˙ 2H3 radical,which subsequently reacts with O2 and produces HO˙ 2 radical.The importance of these species has been explained earlier. Thereactions

have positive sensitivities due to the fact that they compete withthe formation of the cyclohexyl radical.

Because the following reactions

consume reactive radical species and produce stable productsand resonantly stabilized radicals, they also have positivesensitivity coefficients.

Theâ-scission reactions of 1C7H13-6 and eC7H13-a result inthe formation of 1C˙ 4H7-4 radical:

Because 1C˙ 4H7-4 radical decomposes to produce more than halfof all the C2H3 radical generated, the above reactions increasethe reactivity of the system and have negative sensitivitycoefficients.

The reaction

generates an alkene and a methyl radical and competes withthe formation of 1C˙ 4H7-4 from eC7H13-a, thereby reducing thereactivity of the system, resulting in a positive sensitivitycoefficient for this reaction.

As before, a reaction that leads to the formation of a C˙ 2H3

radical

has a negative sensitivity coefficient.Some other reactions

have large sensitivity coefficients. Analysis of reaction editsdid not produce a clear picture of why these reactions had theobserved sensitivities be they positive or negative. However,their sensitivity coefficients do not appear to be directlyassociated with the reactions themselves but can be attributedto the fact that these reactions are in competition with otherpathways, which ultimately leads to an increase or a decreasein the reactivity of the system.

A sensitivity analysis was also performed for MCH oxidationin the flow reactor at 1815 ppm MCH,φ ) 1.3, 1160 K, and 1atm, Figure 21. The sensitivity coefficient is defined as thepercentage differences between the amount of MCH present at1.375 ms (the time at which 10% fuel is consumed in thebaseline mechanism) in the altered mechanism relative to thebaseline mechanism, Figure 21. A negative sensitivity valuesignals an increase in the amount of MCH consumed. Reactionswith a sensitivity value of less than 0.1% were excluded.

For the flow reactor sensitivity analysis increasing theA-factorfor individual reactions in which H-atom abstracts H-atom fromMCH causes the overall concentration of MCH present to fallin some cases and to increase in others, Figure 21. This is dueto the stability of the decomposition products produced by thebreakdown of the different C˙ 7H13 radicals. However, it shouldbe noted that the nature of the C˙ 7H13 isomer formed does notchange the sign of the sensitivity coefficient when a H-atom isabstracted by O˙ H, O, CH3O, or CH3 radicals. Increasing theA-factor for all MCH + CH3 f MCH-R′ + CH4 causes anincrease in the amount of MCH consumed. Conversely,increasing theA-factor for all H-atom abstractions from MCHby OH, O, CH3O causes adecreasein the amount of MCHconsumed. In general, it seems that increasing theA-factor ofreactions that compete with the MCHf CYCHEXRAD + CH3

reaction causes the system to be less reactive, resulting in alower overall consumption of MCH. However, for somereactions, in particular some hydrogen atom abstractions fromMCH by H-atoms andall hydrogen atom abstractions fromMCH by CH3 radicals, the extra reactivity due to increasingtheA-factor more than compensates for the fact that less MCHdecomposes to give cyclohexyl radical and a C˙ H3 radical.

For example, increasing theA-factor of H-atom abstractionby H-atom producing MCH-R1 or MCH-R4 causes an increasein the amount of MCH consumed at 1.375 ms. This occursbecause MCH-R1 breaks down to produce one H-atom and otherstable species, so there is no overall decrease in H-atomconcentration. The same trend is observed when MCH-R4 isproduced. 85% of the MCH-R4 goes to 1C˙ 4H7-4, which gives1,3C4H6 and a H-atom. Another 5% goes to C3H6 and a H-atom.In summary, 90( 7% of MCH-R4 that breaks down via thispath produces one H-atom, keeping the overall H-atom supplyalmost constant.

Conversely, increasing the rates of H-atom abstractions byH-atom forming CYCHEXCH2, MCH-R2, or MCH-R3 resultsin a decrease in the amount of MCH consumed at 1.375 ms.Of the CYCHEXCH2 formed 60% produces allyl (aC˙ 3H5)radical and two ethylene (C2H4) molecules, and 40% producesbuta-1,3-diene (1,3C4H6), C2H4, and a C˙ H3 radical. A methyl

MCH f CYCHEXRAD + CH3

MCH + H f MCH-R′ + H2

(MCHR′ ) all methylcyclohexyl radicals)

MCH + OH f MCH-R′ + H2O

(MCHR′ ) all methylcyclohexyl radicals)

MCH + CH3O f MCH-R1 + CH3OH

MCH + O f MCH-R1 + OH


radical is less reactive than a H-atom because only 19% ofmethyl radicals consume the fuel via H-atom abstraction,compared to 89% of the H-atoms produced. Through theseroutes the reactive radical pool decreases. MCH-R2 decomposesto give three main types of products:

1. 8% MCH-R2 gives iC˙ 5H8 and C2H5 and 18% gives1,3C5H8 and C2H5 radical, with C2H5 radical undergoingâ-scission to yield C2H4 and a H-atom. This is the onlymajor MCH-R2 decomposition pathway that would speedup the rate of decomposition of MCH.

2. 43% of MCH-R2 gives 1,3C4H6 and a C˙ H3 radical,which is much less reactive than a H-atom.

3. An additional 9% of MCH-R2 breaks down to formtwo C2H4 molecules and a C˙ 2H3 radical. 91% of C˙ 2H3

reacts with O2 to form different species, with only 9%undergoingâ-scission to produce ethylene and a hydrogenatom.

All other pathways by which MCH-R2 and its subproductsdecomposed yielded stable intermediates such as 1,3C4H6, C2H4,1,3C5H8, and iC5H8, with negligible amounts of H-atom beingformed. In summary, for every one hundred H-atoms used forhydrogen abstraction forming MCH-R2, less than thirteen werereturned to the cycle. 55% of MCH-R3 breaks down viapathways that produce a H-atom. The other pathways producemore stable species. Therefore, increasing the rate of H-atomabstraction by H-atom forming MCH-R3 leads to an overalldecrease in H-atom concentration, as only half the H-atomsconsumed in abstraction from MCH are replaced.

Increasing the rates for H-atom abstraction by O˙ H, CH3O,and O radicals leads to a decrease in the amount of MCHconsumed. These reactions compete with the decomposition ofMCH forming cyclohexyl and a methyl radical. As describedpreviously, because this latter reaction shows a high negativesensitivity coefficient, any reaction that competes with it willshow a positive sensitivity, as observed.

Increasing the rate of hydrogen atom abstraction by C˙ H3

radicals causes MCH consumption to increase. This is becauseH abstraction is competing with the methyl plus methyl radicalrecombination to form ethane. Increasing H abstraction will leadto fuel decomposition plus a stable CH4 molecule, instead of astable C2H6 molecule with no MCH decomposition.

Conclusions

Ignition delay times have been measured for 0.5% MCH plus10.5% O2 (φ ) 0.5), 1.0% MCH plus 10.5% O2 (φ ) 1.0), and1.0% MCH plus 5.25% O2 (φ ) 2.0) mixtures in Ar, Figures 7and 8. These experiments were carried out behind reflectedshock waves at reflected shock pressures of 1.0, 2.0, and 4.0atm across a temperature range of 1200-2200 K. As expected,the ignition delay times for a particular mixture decreased asthe reflected shock pressure was increased from 1.0 to 2.0, and4.0 atm. Increasing the O2 concentration in the mixture from5.25 to 10.5%, while maintaining a constant MCH concentrationof 1.0%, caused a large decrease in the ignition delays timesobserved. On the other hand, increasing the fuel concentrationfrom 0.5 to 1.0%, while keeping the O2 constant at 10.5%,caused a lengthening in the ignition delay times observed. Thischange was about one-third of that observed when the O2

concentration was changed. The experimental conditions studiedpreviously by Hawthorne and Nixon40 were replicated. Theignition delay times measured agreed well with the previouswork, Figure 6.

A detailed kinetic model was created consisting of 190 speciesand 904 reversible reactions. The model predictions are in goodagreement with the shock tube ignition delay times. The generaltrends displayed by the experimental data, such as changes inignition delay times due to different reflected shock temperaturesand pressure as well as different fuel and O2 concentrations,were also reflected by the simulated delay times. The modelwas also used to simulate species profiles produced for MCHoxidation and pyrolysis in a turbulent flow reactor study byZeppieri et al.,39 Figures 10 and 11. In both cases the modelagrees with the profiles of the MCH decomposition productsbuta-1,3-diene, isoprene, ethylene, and propene. There is anunderestimation of the amount of methane produced and thesimulated MCH decomposition profile does not decrease asrapidly as the experimental data. To allow for the experimentaluncertainties of the time-temperature history of the in-flowingreactants, and to give the best overall agreement between thesimulated and experimental species profiles, the simulatedpyrolysis data was time-shifted by-5 ms; no time shifting wascarried out on the oxidation profiles. Some species predictedby the model were not observed under the oxidation experi-mental conditions; in the cases of hydrogen, water, andformaldehyde this may have been due to the experimental setupused to carry out species analysis, Figure 12. Measurablequantities of penta-1,3-diene were predicted by the model, butthis species was not reported in the experiments, Figure 11.

Finally, sensitivity studies show that MCH oxidation is verysensitive to ethylene/vinyl chemistry, indicating that an exten-sively validated detailed chemical kinetic mechanism for thisfuel is vital to accurately simulate MCH oxidation.

Acknowledgment. J.P.O. thanks Enterprise Ireland (SC/2001/313) and Mayo County Council for financial assistance.The assistance of the Environmental Change Institute and theHigher Education Authority (PRTLI Cycle 2) is also gratefullyacknowledged. We thank Dr. William Pitz (Lawrence LivermoreNational Laboratory) for helpful discussions.

References and Notes

(1) Simmie, J. M.Prog. Energy Combust. Sci.2003, 29, 599-634.(2) Goloniva, E. S.; Fyodorov, G. G.Proc. Combust. Inst.1956, 6,

88-96.(3) Bittner, J. D.; Howard, J. B.Proc. Combust. Inst.1981, 18, 1106-

1116.

Figure 21. Sensitivity analysis, to MCH concentration at 1.375 ms,for flow reactor oxidation at 1815 ppm MCH,φ ) 1.3, 1160 K, and 1atm. (MCHR′ ) all methylcyclohexyl radicals.)


(4) Lovell, A. B.; Brezinsky, K.; Glassman, I.Proc. Combust. Inst.1988, 22, 1063-1074.

(5) Emdee, J. L.; Brezinsky, K.; Glassman, I.J. Phys. Chem.1992,96, 2151-2161.

(6) Lindstedt, R. P.; Skevis, G.Combust. Flame1994, 99, 551-561.(7) Zhang, H. Y.; McKinnon, J. T.Combust. Sci. Technol.1995, 107,

261-300.(8) Tan, Y.; Frank, P.Proc. Combust. Inst.1996, 26, 677-684.(9) Davis, S. G.; Wang, H.; Brezinsky, K.; Law, C. K.Proc. Combust.

Inst. 1996, 26, 1025-1033.(10) Chai, Y.; Pfefferle, L. D.Fuel 1998, 77, 313-320.(11) Alzueta, M. U.; Glarborg, P.; Dam-Johansen, K.Int. J. Chem. Kinet.

2000, 32, 498-522.(12) Scho¨bel, A. Experimentelle und numerische Untersuchung des

Benzolabbaus im Stro¨mungsreaktor bei mittleren Temperaturen, 2000 Wiss.Berichte FZKA 6498, Forschungszentrum Karlsruhe.

(13) Ristori, A.; Dagaut, P.; El Bakali, A.; Pengloan, G.; Cathonnet,M. Combust. Sci. Technol.2001, 167, 223-256.

(14) Lindstedt, R. P.; Maurice, L.; Meyer, M.Faraday Discuss.2001,119, 409-432.

(15) Richter, H.; Howard, J. B.Phys. Chem. Chem. Phys.2002, 4, 2038-2055.

(16) Da Costa, I.; Fournet, R.; Billaud, F.; Battin-Leclerc, F.Int. J. Chem.Kinet. 2003, 35, 503-524.

(17) Brezinsky, K.Prog. Energy Combust. Sci.1986, 12, 1-24.(18) Pamidimukkala, K. M.; Kern, R. D.; Patel, M. R.; Wei, H. C.;

Kiefer, J. H.J. Phys. Chem.1987, 91, 2148-2154.(19) Braun-Unkhoff, M.; Frank, P.; Just, Th.Proc. Combust. Inst.1988,

22, 1053-1061.(20) Lindstedt, R. P.; Maurice, L. Q.Combust. Sci. Technol.1996, 120,

119-167.(21) Klotz, S. D.; Brezinsky, K.; Glassman, I.Proc. Combust. Inst.1998,

27, 337-344.(22) Pengloan, G.; Dagaut, P.; Djebaı¨li-Chaumeix, N.; Paillard, C. E.;

Cathonnet, M.Colloq. Combust. Propre, Orle´ans2001, p 9.(23) Dagaut, P.; Pengloan, G.; Ristori, A.Phys. Chem. Chem. Phys.

2002, 4, 1846-1854.(24) Sivaramakrishnan, R.; Tranter, R. S.; Brezinsky, K.; Durgam, S.;

Vasudevan, H.Proc. Joint Mtg. U. S. Sect. Combust. Inst., 3rd2003, A04.(25) Ribaucour, M.; Roubaud, A.; Minetti, R.; Sochet, L. R.Proc.

Combust. Inst.2000, 28, 1701-1707.(26) Roubaud, A.; Minetti, R.; Sochet, L. R.Combust. Flame2000, 121,

535-541.(27) Roubaud, A.; Lemaire, O.; Minetti, R.; Sochet, L. R.Combust.

Flame2000, 123, 561-571.(28) Dagaut, P.; Ristori, A.; El Bakali, A.; Cathonnet, M.Fuel 2002,

81, 173-184.(29) Lemaire, O.; Ribaucour, M.; Carlier, M.; Minetti, R.Combust.

Flame2001, 127, 1971-1980.(30) Ribaucour, M.; Lemaire, O.; Minetti, R.Proc. Combust. Inst.2002,

29, 1303-1310.(31) Dayma, G.; Glaude, P. A.; Fournet, R.; Battin-Leclerc, F.Int. J.

Chem. Kinet.2003, 35 (7), 273-285.(32) http://www.faqs.org/faqs/autos/gasoline-faq/part1/preamble/html.(33) Yu, J.; Eser, S.Ind. Eng. Chem. Res.1995, 34, 404-409.(34) Violi, A.; Yan, S.; Eddings, E. G.; Sarofim, A. F.; Granata, S.;

Faravelli, T.; Ranzi, E.Combust. Sci. Technol.2002, 174, 399-417.(35) Maurice, L.; Edwards, T.; Griffiths, J.Prog. Astronaut. Aeronaut.

2000, 189, 757-822.(36) Tanaka, S.; Ayala, F.; Keck, J. C.; Heywood, J. B.Combust. Flame

2003, 132, 219-239.(37) Owen, K.; Cooley, T.AutomotiVe Fuels Reference Book, 2nd ed.;

Society of Automotive Engineers: Warrendale, PA, 1995; p 831.(38) Brown, T. C.; King, K. D.Int. J. Chem. Kinet.1989, 21, 251-

266.(39) Zeppieri, S.; Brezinsky, K.; Glassman, I.Combust. Flame1997,

108, 266-286.(40) Hawthorn, R. D.; Nixon, A. C.AIAA 1966, 4, 513-520.(41) Granata, S.; Faravelli, T.; Ranzi, E.Combust. Flame2003, 132,

533-544.(42) Ranzi, E.; Gaffuri, P.; Faravelli, T.; Dagaut, P.Combust. Flame

1995, 103, 91-106.(43) Ranzi, E.; Faravelli, T.; Gaffuri, P.; Sogaro, A.; D’Anna, A.; Ciajolo,

A. Combust. Flame1997, 108, 24-42.(44) Ranzi, E.; Dente, M.; Goldaniga, A.; Bozzano, G.; Faravelli, T.

Prog. Energy Combust. Sci.2001, 27, 99-139.

(45) Ben-Dor, G.; Igra, O.; Elperin, T.; Lifhitz, In. A.Handbook of ShockWaVes; Academic Press: New York, 2001; Vol. 3, pp 212-256.

(46) Morley, C. http://www.gaseq.co.uk/.(47) Smith, J. M. Ph.D. Thesis, National University of Ireland, Galway,

2004.(48) Michael, J. V.; Sutherland, J. W.Int. J. Chem. Kinet.1986, 18,

409-436.(49) Horning, D. C.; Davidson, D. F.; Hanson, R. K.Int. Symp. Shock

WaVes, 23rd 2001, paper 5732.(50) Horning, D. C.; Davidson, D. F.; Hanson, R. K.J. Propul. Power

2002, 18, 363-371.(51) Kee, R. J.; Rupley, F. M.; Miller, J. A.; Coltrin, M. E.; Grcar, J.

F.; Meeks, E.; Moffat, H. K.; Lutz, A. E.; Dixon-Lewis, G.; Smooke, M.D.; Warnatz, J.; Evans, G. H.; Larson, R. S.; Mitchell, R. E.; Petzold, L.R.; Reynolds, W. C.; Caracotsios, M.; Stewart, W. E.; Glarborg, P.; Wang,C.; Adigun, O.; Houf, W. G.; Chou, C. P.; Miller, S. F.Chemkin Collection,Release 3.7; Reaction Design, Inc.: San Diego, CA, 2002.

(52) Laskin, A.; Wang, H.; Law, C. K.Int. J. Chem. Kinet.2000, 32,589-614.

(53) OÄ Conaire, M.; Curran, H. J.; Simmie, J. M.; Pitz, W. J.; Westbrook,C. K. Int. J. Chem. Kinet.2004, 36, 603-622.

(54) Curran, H. J.; Gaffuri, P.; Pitz, W. J.; Westbrook, C. K.Combust.Flame1998, 114, 149-177.

(55) Curran, H. J.; Gaffuri, P.; Pitz, W. J.; Westbrook, C. K.Combust.Flame2002, 129, 253-280.

(56) Prosen, E. J.; Johnson, W. H.; Rossini, F. D.J. Res. Natl. Bur.Stand.1946, 37, 51-56.

(57) Beckett, C. W.; Pitzer, K. S.; Spitzer, R.J. Am. Chem. Soc.1947,69, 2488-2495.

(58) Ritter, E. R.; Bozzelli, J. W.Int. J. Chem. Kinet.1991, 23, 767-778.

(59) Benson, S. W.Thermochemical Kinetics; John Wiley and Sons:New York, 1976.

(60) Lay, T.; Bozzelli, J. W.; Dean, A. M.; Ritter, E. R.J. Phys. Chem.1995, 99 (39), 14514-14527.

(61) Sumathi, R.; Green, W. H., Jr.Phys. Chem. Chem. Phys.2003, 5,3402-3417.

(62) O’Neal, H. E.; Benson, S. W.J. Phys. Chem.1967, 71 (9), 2903-2921.

(63) Tsang, W.J. Phys. Chem. Ref. Data1990, 19, 1-68.(64) Westbrook, C. K.; Pitz, W. J.; Curran, H. J.; Boercker, J.; Kunrath,

E. Int. J. Chem. Kinet.2001, 33, 868-877.(65) Curran, H. J.Int. J. Chem. Kinet., in press.(66) Matheu, D. M.; Green, W. H., Jr.; Grenda, J. M.Int. J. Chem.

Kinet. 2003, 35, 95-119.(67) Solomons, T. W. GrahamOrganic Chemistry, 6th ed.; Wiley: New

York, 1996; p 145.(68) Pine, S. H.Organic Chemistry, 5th ed.; McGraw-Hill: New York,

1987; p 547.(69) Orchard, S. E.; Thrush, B. A.Proc. R. Soc. London A1974, 337,

257-274.(70) Alfassi, Z. B.; Benson, S. W.; Golden, D. M.J. Am. Chem. Soc.

1973, 95, 4784-4788.(71) Dean, A. M.J. Phys. Chem.1985, 89, 4600-4608.(72) Weissman, M. A.; Benson, S. W.J. Phys. Chem.1988, 92, 4080-

4084.(73) Nicovich, J. M.; Ravishankara, A. R.J. Phys. Chem.1984, 88,

2534-2541.(74) Lewis, D. K.; Bergmann, J.; Manjoney, R.; Paddock, R.; Kaira, B.

L. J. Phys. Chem.1984, 88, 4112-4116.(75) Hidaka, Y.; Chimori, T.; Shiba, S.; Suga, M.Chem. Phys. Lett.

1984, 111, 181-183.(76) Skinner, G. B.; Rogers, D.; Patel, K. B.Int. J. Chem. Kinet.1981,

13, 481-495.(77) Newman, C. G.; O’Neal, H. E.; Ring, M. A.; Leska, F.; Shipley,

N. Int. J. Chem. Kinet.1979, 11, 1167-1182.(78) Barnard, J. A.; Parrott, T. K.J. Chem. Soc., Faraday Trans.1976,

1:72, 2404-2415.(79) Tsang, W.Int. J. Chem. Kinet.1973, 5, 651-662.(80) Tsang, W.Int. J. Chem. Kinet.1970, 2, 311-323.(81) Fahr, A.; Stein, S. E.Proc. Combust. Inst.1988, 22, 1023-1029.(82) Newcomb, M.Tetrahedron1993, 49, 1151-1176.(83) Brezinsky, K.; Litzinger, T. A.; Glassman, I.Int. J. Chem. Kinet.

1984, 16, 1053-1074.


Experimental and Modeling Study of Methyl Cyclohexane Pyrolysis and Oxidation

Documents

Transcript of Experimental and Modeling Study of Methyl Cyclohexane Pyrolysis and Oxidation